Description:FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION
(See Section 10 and Rule 13)

Title of Invention:
A SYSTEM AND METHOD FOR BLOOD PRESSURE ESTIMATION FROM VIDEO PHOTOPLETHYSMOGRAPHY (PPG) USING A SMARTPHONE

APPLICANT:
CORTIQA HEALTH PRIVATE LIMITED
An Indian entity having address as:
No.542A, 8th Main Road, 4th Block,
Koramangala, Bengaluru, Bangalore, Karnataka (IN) 560034

The following specification particularly describes the invention and the manner in which it is to be performed.

CROSS-REFERENCE TO RELATED APPLICATIONS AND PRIORITY
The present application does not claim priority from any other application.
TECHNICAL FIELD
The present invention relates to the technical field of blood pressure estimation, and in particular blood pressure detection through a mobile phone using video photoplethysmography. More particularly, the present invention relates to methods, systems, and devices for non-invasive blood pressure measurement utilizing video PPG signals captured by a smartphone's camera and its associated light source.
BACKGROUND
The subject matter discussed in this section is intended to introduce the reader to various aspects of art (the relevant technical field or area of knowledge to which the invention pertains), which may be related to various aspects of the present disclosure that are described or claimed below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present disclosure. Accordingly, it should be understood that these statements in this background section are to be read in this light, and not as admissions of prior art. Similarly, a problem mentioned in the background section or associated with the subject matter of the background section should not be assumed to have been previously recognized in the prior art. The subject matter in the background section merely represents different approaches, which in and of themselves may also correspond to implementations of the claimed technology.
Cardiovascular health involves heart and blood vessel function, including blood pressure, cholesterol levels, and cardiac performance. Maintaining optimal cardiovascular health lowers the risk of heart disease and stroke. Regularly monitoring vital signs like blood pressure, heart rate, temperature, respiratory rate, and oxygen saturation levels is crucial for preventive healthcare, providing insights into overall health status. Cardiovascular mortality ranks highest among non-communicable diseases, accounting for nearly 17.9 million deaths annually, with hypertension emerging as a significant contributor to premature mortality globally. Alarmingly, fewer than half of adults (~42%) with hypertension receive timely diagnosis and treatment, highlighting a critical gap in healthcare delivery and disease management. Hypertension, marked by persistent high blood pressure, significantly increases the risk of cardiovascular conditions such as myocardial infarction, stroke, and heart failure. Conversely, hypotension, or abnormally low blood pressure, can lead to insufficient tissue perfusion, manifesting in symptoms like orthostatic hypotension and syncope.
Blood pressure (BP), a crucial physiological parameter modulated by the sympathetic and parasympathetic pathways of the autonomic nervous system, provides insights into the functioning of the human circulatory system. It indirectly reflects factors such as heart pumping efficiency, peripheral vascular resistance, heart rate, arterial elasticity, blood volume, and overall blood health. The BP is typically measured in millimeters of mercury (mmHg) and includes the systolic pressure (during heart contraction) and diastolic pressure (during heart relaxation). Therefore, Blood pressure measurement is necessary in determining an individual's risk for cardiovascular disease and the need for early treatment. Early detection and treatment of BP may delay or prevent conditions related to high BP, such as stroke. Monitoring blood pressure is crucial for cardiovascular health, and various devices are available for this purpose. These include manual sphygmomanometers, automatic digital monitors, and wearable devices. Manual sphygmomanometers consist of a cuff, a pressure gauge, and a stethoscope, requiring manual inflation and listening for Korotkoff sounds to determine blood pressure readings. However, standardization challenge persists due to variation in cuff size, inflation pressure and calibration methods among manual sphygmomanometers, which contribute to discrepancies in blood pressure readings. Additionally, specialized training or education is required for interpretation of the manual sphygmomanometer. Moreover, due to the static nature of the device, the manual sphygmomanometer possesses limited regulatory oversight and lacks incorporation of population level health data, patient feedback and continuous quality improvement.
Automatic digital monitors are indeed common and user-friendly, typically featuring an inflatable cuff and electronic display. A cuff needs to be placed correctly over user’s arm and inflated. However, they don't always measure blood pressure automatically upon cuff inflation; rather, the user usually initiates the measurement by pressing a button. Additionally, similar to manual sphygmomanometer, standardization challenge persists due to variation in cuff size, inflation pressure and calibration methods among the digital monitors, which contribute to discrepancies in blood pressure readings. Further, cost and affordability of the digital monitoring device, along with recurring expenses for device maintenance and consumables, pose significant barriers to access for individuals with limited financial resources or inadequate health insurance coverage. Moreover, due to the static nature of the device, the automatic digital monitor also possesses limited regulatory oversight and lacks incorporation of population level health data, patient feedback and continuous quality improvement.
Wearable devices such as smartwatches and fitness trackers are indeed gaining popularity for continuous blood pressure monitoring using optical sensors. However, it's worth mentioning that the accuracy and reliability of these devices for blood pressure measurement are still being validated and may vary among different products.
The utilization of an LED flash for photoplethysmography (PPG) to record pulsatile changes in blood volume is a well-known phenomenon, commonly used in devices like pulse oximeters. However, a significant issue with existing PPG devices is their susceptibility to motion artefacts during blood pressure measurement. Even slight movements or muscle contractions can introduce noise and distortions into the PPG waveform, leading to inaccurate readings. For instance, a user may inadvertently shift their arm or hand during measurement, causing fluctuations in the PPG signal that are unrelated to changes in blood volume.
Consider a scenario where a patient is using a wrist-worn PPG device to monitor their blood pressure at home. While taking a measurement, the patient receives a notification on their smartphone, prompting them to glance at the screen and briefly move their arm. This momentary movement introduces motion artefacts into the PPG waveform, affecting the accuracy of the blood pressure reading. As a result, the measured blood pressure may be falsely elevated or diminished, leading to erroneous clinical decisions or unnecessary interventions. Despite technological advancements, algorithmic ambiguity persists in certain cardiovascular monitoring devices, specifically wrist-worn PPG devices, leading to discrepancies in blood pressure readings. This ambiguity could arise from inconsistencies in signal processing algorithms or sensor inaccuracies, resulting in erroneous clinical interpretations and potentially compromising patient care. Additionally, a substantial challenge in wearable devices utilizing optical sensors for blood pressure monitoring may be susceptible to interference from ambient light sources or motion artifact, leading to inaccurate readings.
Addressing the challenge of motion artefacts, ambient light interference and algorithmic ambiguity is crucial for improving the reliability and accuracy of PPG-based blood pressure estimation methods. Advanced signal processing techniques and motion artefact detection algorithms, coupled with machine learning algorithms, can help mitigate these issues and enhance the clinical utility of PPG devices for blood pressure monitoring.
The need for advanced algorithms and signal processing techniques for precise estimation of blood pressure is accurately highlighted, along with the challenge of implementing these techniques in real-time on everyday devices while ensuring accuracy and efficiency.
It's important to clarify that the issues mentioned in the context of traditional blood pressure estimation methods are not limited to accuracy discrepancies and reliability concerns; factors such as cuff size, placement, and patient positioning also play significant roles in the accuracy of blood pressure measurements.
In view of the above, addressing the aforementioned technical challenges requires an improved method for estimating blood pressure.
Further limitations and disadvantages of conventional and traditional approaches will become apparent to one of skill in the art, through the comparison of described systems with some aspects of the present disclosure, as set forth in the remainder of the present application and with reference to the drawings.
SUMMARY
This summary is provided to introduce a Blood pressure estimation system and method using video photoplethysmography (PPG) from a smartphone, and the present discloser is further described below in the detailed description. This summary is not intended to identify essential features of the claimed subject matter nor is it intended for use in classifying or limiting the scope of the claimed subject matter.
According to embodiments illustrated herein, there is provided a method for blood pressure estimation from video photoplethysmography (PPG) using one or more machine learning (ML) techniques. The method may be implemented by an electronic device including one or more processors and a memory communicatively coupled to the processor and the memory is configured to store processor-executable programmed instructions. The method for blood pressure estimation from video photoplethysmography (PPG) using machine learning techniques may involve a series of steps. First, a video of the user's fingertip is captured by one or more optical sensors on a portable device along with a light source. Further, the method may comprise the step of extracting a red channel from each frame of one or more frames of the received video. Further, the method may comprise a step of obtaining a PPG waveform by taking an average of one or more red channels from the one or more frames of the received video. Further, the method may comprise a step of inverting the PPG waveform to correct inversion format of the PPG waveform and obtain an inverted PPG waveform. Further, the method may comprise a step of filtering the inverted PPG waveform to remove noise from the inverted PPG waveform and obtain a filtered PPG waveform. Further, the method may comprise one or more steps of selecting the filtered PPG waveform based on detecting signal strength and signal quality. Further, the method may comprise a step for extracting a portion of PPG pulse from the selected PPG waveform. Further, the method may comprise a plurality of steps for performing the pulse correction on the PPG pulse. The plurality of steps for performing the pulse correction may comprise a step of detecting peaks and troughs from the PPG pulse. The peak may be the highest point between two troughs. Further, the plurality of steps for performing the pulse correction may correspond to detecting an ascending section and a descending section of the PPG pulse. The ascending section may be starting from the first trough of the PPG pulse ending at the peak of the PPG pulse and the descending section may be starting from the peak of the PPG pulse ending at the second trough of the PPG pulse. Further, the plurality of steps for performing the pulse correction may comprise a step of calculating areas of both the ascending section and the descending section of the PPG pulse. Further, the plurality of steps for performing the pulse correction may comprise a step of flipping the PPG pulse if area of the ascending section may be greater than the area of the descending section. Further, the method may comprise one or more steps for validating the PPG pulse through fiducial point detection and by defining thresholds on the fiducial points. Further the method may comprise step for extracting one or more features from the validated PPG pulse. Finally, these features may be applied to one of a systolic regression model, a diastolic regression model, or a combination of both, to estimate the user's blood pressure. Furthermore, the method may integrate video-based PPG technology with machine learning algorithms to provide a non-invasive approach to blood pressure monitoring using readily accessible portable devices.
According to embodiments illustrated herein, there is provided a system for blood pressure estimation from video photoplethysmography (PPG) from a smartphone using one or more machine learning (ML) techniques is disclosed. The system may comprise one or more portable devices. Further, the one or more portable devices may be coupled with one or more optical sensors and a light source. Further, the system may comprise a memory and a processor. Further, the processor may be configured to execute programmed instructions stored in the memory. Further, the processor may be configured to perform a step of receiving a video of user’s fingertip using one or more optical sensors coupled with the one or more portable devices along with the light source. Further, the processor may process a video of a user's fingertip captured by the optical sensors. Further, the processor may be configured for extracting a red channel from each frame of one or more frames of the received video. Further, the processor may be configured for obtaining a PPG waveform by taking average of one or more red channels from one or more frames of the received video. Further, the processor may be configured for inverting the PPG waveform to correct inversion format of the PPG waveform and obtain an inverted PPG waveform. Further, the processor may be configured for filtering the inverted PPG waveform to remove noise from the inverted PPG waveform and obtain an inverted PPG waveform. Further, the processor may be configured for selecting the filtered PPG waveform based on detecting signal strength and signal quality. Further, the processor may be configured for extracting a portion of PPG pulse from the selected PPG waveform. Further, the processor may be configured to perform one or more steps for performing pulse correction on the PPG pulse. The pulse correction may be performed in one or more steps. Further, one or more steps may correspond to detecting peaks and troughs from the PPG pulse. The peak is the highest point between two troughs. Further, the one or more steps may correspond to detecting an ascending section and a descending section of the PPG pulse. The ascending section may be starting from the first trough of the PPG pulse ending at the peak of the PPG pulse and the descending section may be starting from the peak of the PPG pulse ending at a second trough of the PPG pulse. Further, one or more steps may correspond to calculating areas of both the ascending section and the descending section of the PPG pulse. Further, one or more steps may correspond to flipping the PPG pulse if area of the ascending section may be greater than the area of the descending section. Further, the processor may be configured for validating the PPG pulse through fiducial point detection, fiducial point validation by defining thresholds on the fiducial points. Further, the processor may be configured for extracting one or more features from the validated PPG pulse. Further, these features may subsequently be utilized in estimate the user's blood pressure. Furthermore, the processor may be configured for applying one or more features to either a systolic regression model, a diastolic regression model, or a combination of both, to estimate the blood pressure of the user. The system may aim to provide accurate blood pressure estimation using non-invasive video-based PPG technology in conjunction with machine learning algorithms, potentially enabling continuous monitoring of cardiovascular health.
The foregoing summary is illustrative only and is not intended to be in any way limiting. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.
BRIEF DESCRIPTION OF DRAWINGS
The accompanying drawings illustrate the various embodiments of systems, methods, and other aspects of the disclosure. Any person with ordinary skills in art will appreciate that the illustrated element boundaries (e.g., boxes, groups of boxes, or other shapes) in the figures represent one example of the boundaries. In some examples, one element may be designed as multiple elements, or multiple elements may be designed as one element. In some examples, an element shown as an internal component of one element may be implemented as an external component in another, and vice versa. Further, the elements may not be drawn to scale.
Various embodiments will hereinafter be described in accordance with the appended drawings, which are provided to illustrate and not to limit the scope in any manner, wherein similar designations denote similar elements, and in which:
The detailed description is described with reference to the accompanying figures. In the Figures, the left-most digit(s) of a reference number identifies the Figure in which the reference number first appears. The same numbers are used throughout the drawings to refer to the like features and components.
FIG. 1 is a block diagram that illustrates a system environment (100) for blood pressure estimation using video photoplethysmography (PPG), in accordance with an embodiment of present subject matter.
FIG. 2 is a block diagram (100) that illustrates the visual arrangement of capturing video of user’s fingertip for blood pressure estimation using the system (100), in accordance with an embodiment of present subject matter.
FIG. 3 is a block diagram (300) that illustrates components of an application server (104) configured for performing steps for blood pressure estimation using video photoplethysmography (PPG), in accordance with an embodiment of the present subject matter
FIG. 3 A illustrates a block diagram of the input/output unit (304) of the application server (104), in accordance with an embodiment of the present subject matter.
FIG. 3B illustrates a block diagram of pre-processing unit (305) of the application server (104), in accordance with an embodiment of the present subject matter.
FIG. 3C illustrates a block diagram of signal analysis and selection unit (306) of the application server (104), in accordance with an embodiment of the present subject matter.
FIG. 3D illustrates a block diagram of the pulse extraction and validation unit (307) of the application server (104), in accordance with an embodiment of the present subject matter.
FIG. 3E illustrates a block diagram of feature extraction unit (308) of the application server (104), in accordance with an embodiment of the present subject matter.
FIG. 3F illustrates a block diagram of the Blood Pressure Estimation unit (309) of the application server (104), in accordance with an embodiment of the present subject matter.
FIGS. 4A-4B is a flowchart (400) that illustrates a method for blood pressure estimation using video photoplethysmography (PPG), in accordance with an embodiment of the present subject matter; and
FIG. 5 illustrates a block diagram (500) of an exemplary computer system for implementing embodiments consistent with the present subject matter.
DETAILED DESCRIPTION
The present disclosure may be best understood with reference to the detailed figures and description set forth herein. Various embodiments are discussed below with reference to the figures. However, those skilled in the art will readily appreciate that the detailed descriptions given herein with respect to the figures are simply for explanatory purposes as the methods and systems may extend beyond the described embodiments. For example, the teachings presented, and the needs of a particular application may yield multiple alternative and suitable approaches to implement the functionality of any detail described herein. Therefore, any approach may extend beyond the particular implementation choices in the following embodiments described and shown.
The terms “comprise”, “comprising”, “include(s)”, or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a setup, system or method that comprises a list of components or steps does not include only those components or steps but may include other components or steps not expressly listed or inherent to such setup or system or method. In other words, one or more elements in a system or apparatus preceded by “comprises… a” does not, without more constraints, preclude the existence of other elements or additional elements in the system or apparatus.
Reference throughout the specification to “various embodiments,” “some embodiments,” “one embodiment,” or “an embodiment” means that a particular feature, structure, or method described in connection with the embodiment is included in at least one embodiment. Thus, appearances of the phrases “in various embodiments,” “in some embodiments,” “in one embodiment,” or “in an embodiment” in places throughout the specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
The present disclosure relates to a system and method for blood pressure estimation from video photoplethysmography (PPG) using one or more machine learning techniques. The system and method may include the use of a smartphone for capturing video photoplethysmography (PPG). The method is implemented by an electronic device including one or more processors and a memory communicatively coupled to the processor and the memory is configured to store processor-executable instructions. The method includes receiving a video of user’s fingertip. The video is captured using one or more optical sensors coupled with the electronic device along with a light source. The method includes extracting a red channel from each frame of one or more frames of the received video. Further, the method includes obtaining a PPF waveform by taking an average of one or more red channels from one or more frames of the received video. Further, the method includes inverting the PPG waveform to correct inversion format of the PPG waveform and obtain an inverted PPG waveform. Further, the method includes filtering the inverted PPG waveform to remove noise from the inverted PPG waveform and obtain a filtered PPG waveform. Further, the method includes selecting the filtered PPG waveform based on detecting signal strength and signal quality. Further, the method includes extracting a portion of PPG pulse from the selected PPG waveform. Further, the method includes performing a pulse correction on the PPG pulse. The pulse correction is performed by the following steps. Further, the method includes detecting peaks and troughs from the PPG pulse. A peak is the highest point between two troughs. Further, the method includes detecting an ascending section and a descending section of the PPG pulse. The ascending section starts from a first trough of the PPG pulse, ending at the peak of the PPG pulse and the descending section is starting from the peak of the PPG pulse, and ends at a second trough of the PPG pulse. Further, the method includes calculating areas of both the ascending section and the descending section of the PPG pulse. Further, the method includes flipping the PPG pulse if area of the ascending section is greater than the area of the descending section. Further, the method includes validating the PPG pulse through fiducial point detection and fiducial point validation by defining thresholds on the fiducial points. Further, the method includes extracting one or more features from the validated PPG pulse. Further the method includes applying one or more extracted features to one of a systolic regression model, a diastolic regression model, and a combination thereof, to estimate the blood pressure of the user.
The objective of the present disclosure is to develop a non-invasive method for blood pressure estimation using video PPG technology, aiming to eliminate the need for traditional cuff-based measurements and its associated discomfort and risk. This method aims to leverage (PPG) technology to estimate blood pressure without the requirement for direct physical contact with the patient's skin.
Another objective of the present disclosure is to ensure accurate blood pressure estimation using video PPG by implementing robust calibration and validation procedures.
Yet another objective of the present disclosure is to overcome the significant issues of inadvertent motion artifacts and ambient light interference during blood pressure measurement in existing PPG devices. The present disclosure describes various signal processing methodologies for video signals from the smartphone which will effectively remove the non-performing/erroneous PPG signals and identify/adapting the important PPG signals for blood pressure estimation.
Yet another objective of the present disclosure is to implement machine learning model and signal processing techniques for continuous, real-time and accurate blood pressure estimation. This real-time monitoring capability is valuable for tracking changes in blood pressure and detecting abnormalities promptly.
Yet another objective of the present disclosure is to integrate seamlessly with smartphones and other smart devices, enabling integration with existing digital health platforms and applications. This facilitates personalized health management and data sharing for healthcare professionals. Integrating with existing smartphones provides cost effective solutions to the costly and affordability barrier of conventional healthcare monitoring devices, specifically for individuals with limited financial resources or inadequate health insurance coverage.
Yet another objective of the present disclosure is to facilitate continues improvement performance and accuracy over time. The present disclosure incorporates population level health data and includes dynamic features of patient feedback and continuous quality improvement. Due to dynamic nature of the present disclosure, the blood pressure monitored through the disclosed method, always align with the current regulatory healthcare standards to overcome standardization challenge of traditional monitoring device.
Yet another objective of the present disclosure is to offer a user-friendly and intuitive method for measuring blood pressure, ensuring ease of use and accessibility for all users, and convenient blood pressure monitoring anytime and anywhere, without the need for additional specialized equipment or corresponding training and education. The present disclosure enhances accessibility to blood pressure monitoring, especially in remote or underserved areas where access to medical facilities may be limited.
Yet another objective of the present disclosure is to promote widespread adoption of video PPG-based blood pressure estimation by addressing user concerns, optimizing device performance, and providing comprehensive training and support resources.
Yet another objective of the present disclosure is to encompass the overarching goal of improving blood pressure estimation methods to enhance the overall experience for individuals with hypertension or other cardiovascular conditions. It emphasizes the importance of accuracy, reliability, user comfort, convenience, and actionable insights in the development of blood pressure monitoring technologies.
FIG. 1 is a block diagram that illustrates a system environment (100) for blood pressure estimation using video photoplethysmography (PPG), in accordance with an embodiment of present subject matter. The system environment (100) typically includes a database server (102), an application server (104), a communication network (106), and one or more portable devices (108). The database server (102), the application server (104), and one or more portable devices (108) are typically communicatively coupled with each other via the communication network (106). In an embodiment, the application server (104) may communicate with the database server (102), and the one or more portable devices (108) using one or more protocols such as, but not limited to, Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol (WAP), RF mesh, Bluetooth Low Energy (BLE), and the like, to communicate with one another.
In an embodiment, the database server (102) may refer to a computing device that may be configured to store user’s fingertip data, training dataset, other intermediate processing data. In an embodiment, the database server (102) may include a special purpose operating system specifically configured to perform one or more database operations on the stored content. Examples of database operations may include, but are not limited to, Select, Insert, Update, and Delete. In an embodiment, the database server (102) may include hardware that may be configured to perform one or more predetermined operations. In an embodiment, the database server (102) may be realized through various technologies such as, but not limited to, Microsoft® SQL Server, Oracle®, IBM DB2®, Microsoft Access®, PostgreSQL®, MySQL®, SQLite®, distributed database technology and the like. In an embodiment, the database server (102) may be configured to utilize the application server (104) for storage and retrieval of data used for securely performing one or more operations in a user interface platform.
A person with ordinary skills in art will understand that the scope of the disclosure is not limited to the database server (102) as a separate entity. In an embodiment, the functionalities of the database server (102) can be integrated into the application server (104) or into one or more portable device (108).
In an embodiment, the application server (104) may refer to a computing device or a software framework hosting an application or a software service. In an embodiment, the application server (104) may be implemented to execute procedures such as, but not limited to, programs, routines, or scripts stored in one or more memories for supporting the hosted application or the software service. In an embodiment, the hosted application or the software service may be configured to perform one or more predetermined operations. The application server (104) may be realized through various types of application servers such as, but are not limited to, a Java application server, a .NET framework application server, a Base4 application server, a PHP framework application server, or any other application server framework.
In an embodiment, the application server (104) may be configured to utilize the database server (102) and the one or more portable device (108), in conjunction, for securely performing one or more operations in a user interface platform. In an implementation, the application server (104) corresponds to the user interface platform for securely performing one or more operations recommended by one or more users.
In an embodiment, the communication network (106) may correspond to a communication medium through which the application server (104), the database server (102), and the one or more portable device (108) may communicate with each other. Such a communication may be performed in accordance with various wired and wireless communication protocols. Examples of such wired and wireless communication protocols include, but are not limited to, Transmission Control Protocol and Internet Protocol (TCP/IP), User Datagram Protocol (UDP), Hypertext Transfer Protocol (HTTP), Wireless Application Protocol (WAP), File Transfer Protocol (FTP), ZigBee, EDGE, infrared IR), IEEE 802.11, 802.16, 2G, 3G, 4G, 5G, 6G, 7G cellular communication protocols, and/or Bluetooth (BT) communication protocols. The communication network (106) may either be a dedicated network or a shared network. Further, the communication network (106) may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, and the like. The communication network (106) may include, but is not limited to, the Internet, intranet, a cloud network, a Wireless Fidelity (Wi-Fi) network, a Wireless Local Area Network (WLAN), a Local Area Network (LAN), a cable network, the wireless network, a telephone network (e.g., Analog, Digital, POTS, PSTN, ISDN, xDSL), a telephone line (POTS), a Metropolitan Area Network (MAN), an electronic positioning network, an X.25 network, an optical network (e.g., PON), a satellite network (e.g., VSAT), a packet-switched network, a circuit-switched network, a public network, a private network, and/or other wired or wireless communications network configured to carry data.
In an embodiment, the one or more portable devices (108) may refer to a computing device used by a user. The one or more portable devices (108) may comprise of one or more processors and one or more memory. The one or more memories may include computer readable code that may be executable by one or more processors to perform predetermined operations. In an embodiment, the one or more portable devices (108) may present a web user interface to transmit the training file and the testing file to the application server (102). Example web user interfaces presented on the one or more portable devices (108) to display a portal visualizing user profile and relevant information, to securely perform one or more operations in the user interface platform. Examples of one or more portable devices (108) may include, but are not limited to, a personal computer, a laptop, a personal digital assistant (PDA), a mobile device, a tablet, or any other computing device.
The system (100) can be implemented using hardware, software, or a combination of both, which includes using where suitable, one or more computer programs, mobile applications, or “apps” by deploying either on-premises over the corresponding computing terminals or virtually over cloud infrastructure. The system (100) may include various micro-services or groups of independent computer programs which can act independently in collaboration with other micro-services. The system (100) may also interact with a third-party or external computer system. Internally, the system (100) may be the central processor of all requests for transactions by the various actors or users of the system. A critical attribute of the system (100) is that it can concurrently and instantly complete an online transaction by a system user in collaboration with other systems. In a specific embodiment, the system (100) is implemented to securely perform one or more operations in the user interface platform.
Now referring to FIG. 2, a block diagram (200) showing visual arrangement of capturing video of user’s fingertip for blood pressure estimation using the system (100), is illustrated in accordance with an embodiment of present subject matter. The block diagram (200) comprises a user’s fingertip (201), one or more optical sensors (202), one or more portable devices (203), and a light source (205). In one embodiment, the one or more optical sensors (202) and the light source (205) may be integrated within the one or portable device (203). In another embodiment, the one or more optical sensors (202), the light source (205) and the one or more portable device (203) may be communicatively coupled and may operate in conjunction with each other. The one or more optical sensors (202) may be configured to capture video of user’s fingertip. The light source (205) may be configured to provide light to surrounding objects. In an exemplary embodiment, user may need to place the user’s fingertip (201) on the one or more optical sensor (202), the light source (205) may illuminate the user’s fingertip (201) placed on the one or more optical sensor (202). Then one or more optical sensors (202) may be configured to capture video of the user’s fingertip (201), illuminated by the light source (205), for a predetermined time period. The one or more portable devices (203) are configured to receive the video of the user’s fingertip (201) captured by the one or more optical sensors (202). Further, the one or more portable devices (203) may comprise one selected from a group consisting of a smartphone, IoT device, personal digital assistant (PDA), laptop computer, stationary personal computer, IPTV remote control, web tablet, laptop computer, pocket PC, a television set capable of receiving IP based video services and mobile IP device. Further, one or more optical sensors (202) may correspond to rear camera of the smartphone. Further, the light source (205) corresponds to a flashlight placed on rear side of the smartphone and generally placed adjacent to the optical sensor (202).
FIG. 3 illustrates a block diagram (300) illustrating components of the application server (104) configured for performing stepwise blood pressure estimation using video photoplethysmography (PPG), in accordance with an embodiment of present invention. Further, FIG. 3 is explained in conjunction with elements from FIG. 1. Here, the application server (104) preferably includes a processor (301), a memory (302), a transceiver (303), an input/output unit (304), a pre-processing unit (305), a signal analysis and selection unit (306), a pulse extraction and validation unit (307), a feature extraction unit (308), a blood pressure estimation unit (309). The processor (301) is further preferably communicatively coupled to the memory (302), the transceiver (303), the input/output unit (304), the pre-processing unit (305), the signal analysis and selection unit (306), the pulse extraction and validation unit (307), the feature extraction unit (308), the blood pressure estimation unit (309), while the transceiver (306) is preferably communicatively coupled to the communication network (106).(not shown)

The processor (301) comprises suitable logic, circuitry, interfaces, and/or code that may be configured to execute a set of instructions stored in the memory (302), and may be implemented based on several processor technologies known in the art. The processor (301) works in coordination with the transceiver (303), the input/output unit (304), the pre-processing unit (305), the signal analysis and selection unit (306), the pulse extraction and validation unit (307), the feature extraction unit (308), the blood pressure estimation unit (309) for blood pressure estimation from video photoplethysmography (PPG). Examples of the processor (301) include, but not limited to, standard microprocessor, microcontroller, central processing unit (CPU), an X86-based processor, a Reduced Instruction Set Computing (RISC) processor, an Application- Specific Integrated Circuit (ASIC) processor, and a Complex Instruction Set Computing (CISC) processor, distributed or cloud processing unit, state machines, logic circuitries, and/or any devices that manipulate signals based on operational instructions and/or other processing logic that accommodates the requirements of the present invention.
The memory (302) comprises suitable logic, circuitry, interfaces, and/or code that may be configured to store the set of instructions, which are executed by the processor (301). Preferably, the memory (302) is configured to store one or more programs, routines, or scripts that are executed in coordination with the processor (301). Additionally, the memory (302) may include any computer-readable medium or computer program product known in the art including, for example, volatile memory, such as static random-access memory (SRAM) and dynamic random-access memory (DRAM), and/or non-volatile memory, such as read only memory (ROM), erasable programmable ROM, a Hard Disk Drive (HDD), flash memories, Secure Digital (SD) card, Solid State Disks (SSD), optical disks, magnetic tapes, memory cards, virtual memory and distributed cloud storage. The memory (302) may be removable, non-removable, or a combination thereof. Further, the memory (302) may include routines, programs, objects, components, data structures, etc., which perform particular tasks or implement particular abstract data types. The memory (302) may include programs or coded instructions that supplement applications and functions of the system (100). In one embodiment, the memory (302), amongst other things, serves as a repository for storing data processed, received, and generated by one or more of the programs or the coded instructions. In yet another embodiment, the memory (302) may be managed under a federated structure that enables adaptability and responsiveness of the application server (104).
The transceiver (303) comprises suitable logic, circuitry, interfaces, and/or code that may be configured to receive, process or transmit information, data or signals, which are stored by the memory (302) and executed by the processor (301). The transceiver (303) is preferably configured to receive, process or transmit, one or more programs, routines, or scripts that are executed in coordination with the processor (301). The transceiver (303) is preferably communicatively coupled to the communication network (106) of the system (100) for communicating all the information, data, signal, programs, routines or scripts through the network.
The transceiver (303) may implement one or more known technologies to support wired or wireless communication with the communication network (106). In an embodiment, the transceiver (303) may include, but is not limited to, an antenna, a radio frequency (RF) transceiver, one or more amplifiers, a tuner, one or more oscillators, a digital signal processor, a Universal Serial Bus (USB) device, a coder-decoder (CODEC) chipset, a subscriber identity module (SIM) card, and/or a local buffer. Also, the transceiver (303) may communicate via wireless communication with networks, such as the Internet, an Intranet and/or a wireless network, such as a cellular telephone network, a wireless local area network (LAN) and/or a metropolitan area network (MAN). Accordingly, the wireless communication may use any of a plurality of communication standards, protocols and technologies, such as: Global System for Mobile Communications (GSM), Enhanced Data GSM Environment (EDGE), wideband code division multiple access (W-CDMA), code division multiple access (CDMA), time division multiple access (TDMA), Bluetooth, Wireless Fidelity (Wi-Fi) (e.g., IEEE 802.11a, IEEE 802.11b, IEEE 802.11g and/or IEEE 802.11n), voice over Internet Protocol (VoIP), Wi-MAX, a protocol for email, instant messaging, and/or Short Message Service (SMS).
The input/output (I/O) unit (304) comprises suitable logic, circuitry, interfaces, and/or code that may be configured to receive or present information. The input/output unit (304) comprises of various input and output devices that are configured to communicate with the processor (301). Examples of the input devices include, but are not limited to, a keyboard, a mouse, a joystick, a touch screen, a microphone, a camera, and/or a docking station. Examples of the output devices include, but are not limited to, a display screen and/or a speaker. The I/O unit (304) may include a variety of software and hardware interfaces, for example, a web interface, a graphical user interface, and the like. The I/O unit (304) may allow the system (100) to interact with the user directly or through the portable devices (108). Further, the I/O unit (304) may enable the system (100) to communicate with other computing devices, such as web servers and external data servers (not shown). The I/O unit (304) can facilitate multiple communications within a wide variety of networks and protocol types, including wired networks, for example, LAN, cable, etc., and wireless networks, such as WLAN, cellular, or satellite. The I/O unit (308) may include one or more ports for connecting a number of devices to one another or to another server. In one embodiment, the I/O unit (304) allows the application server (104) to be logically coupled to other portable devices (108), some of which may be built in. Illustrative components include tablets, mobile phones, wireless devices, etc.
FIG.3A illustrates input/output (I/O) unit (304) of the application server (104), in accordance with an embodiment of the present subject matter. Further, the input/output (I/O) unit (304) comprises a step of video recording (310) and a step for extracting frames (311). Further, the data collected by the input/output (I/O) unit (304) may include the video recording (310) of the user’s fingertip by placing the fingertip on the one or more optical sensors (202) of the one or more portable device (203). Further, the input/output (I/O) unit (304) is configured for extracting (311) one or more frames from the captured video.
Further, FIG. 3B illustrates the preprocessing unit (305) of the application server (104), in accordance with an embodiment of the present subject matter. The preprocessing unit (305) may be configured to extract red channels (312) from each frame of the one or more frames of the captured video of the user’s fingertip (201). After red channel extraction from each frame of the one or more frames, the preprocessing unit (305) may be configured to obtain a PPG waveform from the one or more red channels extracted from the one or more frames of the captured video. In an embodiment, the PPG waveform may be obtained by taking an average of one or more red channels from the one or more frames of the captured video. Further, the preprocessing unit (305) may be configured for inverting the PPG waveform (313) to correct inversion format of the PPG waveform and obtain an inverted PPG waveform. In one operation, inverting the PPG waveform corresponds to vertically inverting the PPG waveform. Further, inverting the PPG waveform corresponds to the inverting polarity of the PPG waveform. Further, the preprocessing unit (305) may be configured for filtering of the inverted PPG waveform to remove noise from the inverted PPG waveform and obtain a filtered PPG waveform. The filtering of the inverted PPG waveform may be performed using one or more filtering techniques. Further, one or more filtering techniques may correspond to a bandpass filter and butter worth filter (314), moving average filtering, signal smoothing, signal interpolation and a combination thereof.
In operation, the present invention may be implemented within a pre-processing unit (305), wherein various filtering techniques may be applied to the PPG waveform. One potential technique involves utilizing a Bandpass filter (314) that may selectively pass frequencies within a specified range relevant to the PPG signal, potentially attenuating frequencies outside this range. This targeted filtering approach may effectively reduce noise that could interfere with the clarity of the PPG waveform.
In another embodiment of the present invention, a Butterworth filter (314) may be integrated into the pre-processing unit (305) to potentially further refine the filtered PPG waveform. This may involve filtering the PPG waveform using one or more specified filtering techniques. These filtering techniques may include bandpass filtering, signal smoothing, and signal interpolation, or a combination thereof. The Butterworth filter, known for its smooth frequency response, may be configured to suppress unwanted noise while preserving the essential characteristics of the PPG signal, including its shape and temporal dynamics.
In yet another embodiment of the present invention, the combined use of Bandpass and Butterworth filters (314) within the pre-processing unit (305) may potentially contribute to comprehensive noise reduction, resulting in potentially enhanced PPG waveforms suitable for accurate cardiovascular analysis. By aiming to improve the signal-to-noise ratio of the PPG data, this method may offer potential benefits in extracting clinically relevant information from video-derived PPG waveforms, thereby advancing the potential applications in non-invasive cardiovascular monitoring.
FIG. 3C illustrates the signal analysis and selection unit (306) of the application server (104), in accordance with an embodiment of the present subject matter. The signal analysis and selection unit (306) may be configured for detecting the signal strength and signal quality. Further, detects a signal from the pre-processed data, and it may detect signal strength by quantifying noise present in the signal. Further, the detected signal may be a PPG signal.
In operation, Referring to Figure.3C, a signal analysis, and selection unit (306) is illustrated, in accordance with an embodiment of the present subject matter. The signal analysis, and selection unit (306) may include various techniques such as thresholding (315), kurtosis (316), baseline wander removal and entropy (317) to assess the signal quality and select the PPG waveform. Further signal analysis, and selection unit (306) analyse signal quality and select the signal strength from the PPG waveform. In another embodiment of the present invention, the signal analysis, and selection unit (306) may perform thresholding (315) to determine specific thresholds for signal acceptance based on predefined criteria. This technique helps identify segments of the PPG waveform that meet certain amplitude or noise level requirements. Furthermore, kurtosis (316) may be utilized to evaluate the shape and distribution of the PPG waveform, providing insights into the presence of outliers or the sharpness of peaks, which can indicate signal quality. Furthermore, the entropy (317) analysis within the signal analysis, and selection unit (306) assesses the complexity and randomness of the PPG waveform, aiding in the identification of noise patterns and overall signal stability.
In another embodiment of the present invention, the signal analysis, and selection unit (306) may analyze signal quality and select the signal strength from the PPG waveform. In operation, the signal analysis, and selection unit (306) may be configured for detecting the signal strength of the filtered PPG waveform by quantifying noise present in the filtered PPG waveform.
FIG. 3D illustrates the pulse extraction and validation unit (307) of the application server (104), in accordance with an embodiment of the present subject matter. In operation, the pulse extraction and validation unit (307) is configured for pulse detection (318). Further, the pulse detection (318) may be configured for extracting a portion of the PPG pulse from the selected PPG waveform. In this process, the extraction of the PPG pulse portion from the waveform can be accomplished by sliding a window of a predetermined time period over the PPG waveform without overlapping. During the PPG pulse extraction, the sliding window technique enables targeted segmentation of the PPG waveform, facilitating the isolation of individual PPG pulse segments. By sliding the window over the PPG waveform at predetermined intervals, each window captures a distinct portion of the PPG signal, aligning with the temporal characteristics of pulsatile events. Further, the pulse extraction and validation unit (307) is configured for performing a pulse correction (321) on the PPG pulse. Further, the pulse correction (321) is performed by executing one or more steps mentioned below. Further, the pulse correction and validation unit (307) is configured for detecting peaks and troughs from the PPG pulse. A peak is the highest point between two troughs. Further, the pulse correction and validation unit (307) is configured for detecting an ascending section and a descending section of the PPG pulse. Further, the ascending section starts from the first trough of the PPG pulse ending at the peak of the PPG pulse and the descending section starts from the peak of the PPG pulse ending at a second trough of the PPG pulse. Further, the pulse correction and validation unit (307) is configured for calculating the areas of both the ascending section and the descending section. Further, the ascending section corresponds to the start of the PPG pulse to the peak of the PPG pulse. Further, the descending section corresponds to the peak of the PPG pulse to the end of the PPG pulse. Further, the pulse correction and validation unit (307) may be configured to compare (319) the area of the ascending section with the area of the descending section. The pulse correction and validation unit (307) may be configured to perform the PPG pulse correction, to refine their quality for accurate cardiovascular analysis, if the area of the ascending section is greater than the area of the descending section. The pulse correction and validation unit (307) may be configured to proceed for nothing to do with the pulse (320) if the area of the ascending section is not greater than the area of the descending section. The correction of the PPG pulse may correspond to flipping the PPG pulse. Further, flipping of the pulse corresponds to horizontally flipping the PPG pulse. Further, flipping of the pulse corresponds to reversing the phase of the PPG pulse.
Further, the pulse correction and validation unit (307) may be configured to send the PPG pulse for pulse validation (322). In an embodiment, the pulse validation may be performed through fiducial point detection, fiducial point validation by defining thresholds on the fiducial points. The fiducial points of the PPG pulse comprise the start point of the PPG pulse, maximum slope point, systolic peak, inflection point, dicrotic notch, diastolic peak, and end point of the PPG pulse. Further, fiducial points of a velocity plethysmogram (VPG) pulse comprises u-peak, v-peak, and w-peak. The VPG pulse is the first derivative of the PPG pulse. Further, fiducial points of an acceleration plethysmogram (APG) pulse comprises a-peak, b-peak, c-peak, d-peak, and e-peak. The APG pulse is a second derivative of the PPG pulse.In one embodiment, detection of the fiducial points of the PPG pulse by the pulse extraction and validation unit (307) is performed by finding the start point and the end point of the PPG pulse, wherein the start point, and the end point of the PPG pulse are local minima of the PPG pulse. Further, detection of the fiducial points of the PPG pulse by the pulse extraction and validation unit (307) is performed by finding the maximum slope point of the PPG pulse based on the maxima of the VPG pulse, wherein the maxima of the VPG pulse corresponds to the u-peak of the VPG pulse. Further, detection of the fiducial points of the PPG pulse by the pulse extraction and validation unit (307) is performed by finding the diastolic peak of the PPG pulse based on the minima of the APG pulse, wherein the minima of the APG pulse corresponds to the e-peak of the APG pulse. Further, detection of the fiducial points of the PPG pulse by the pulse extraction and validation unit (307) is performed by finding the dicrotic notch of the PPG pulse based on coinciding minima point of the VPG pulse with the PPG pulse. Further, the minima point of the VPG pulse corresponds to the v-peak of the VPG pulse. Further, detection of the fiducial points of the PPG pulse by the pulse extraction and validation unit (307) is performed by finding the inflection point of the PPG pulse based on a peak of the APG pulse close to a value zero, wherein the peak of the APG pulse close to zero corresponds to one of the c-peak, d-peak or a combination thereof.
Furthermore, following fiducial point detection, the pulse extraction and validation unit (307)may perform validation of these detected fiducial points through defining thresholds and rules. For example, the last minima of the APG pulse may be considered as the diastolic peak of the PPG pulse if multiple minima are present in the APG pulse. Further, the diastolic peak of the PPG pulse should not be equal to either the maximum slope point or the systolic peak of the PPG pulse. Further, the dicrotic notch of the PPG pulse may be defined as the local minima of either the APG or VPG pulse if the minima point of the VPG pulse does not coincide with the PPG pulse, and it must lie between the systolic peak of the PPG pulse and diastolic peak of the PPG pulse. Furthermore, the inflection point of the PPG pulse may be calculated as the average of the diastolic peak of the PPG pulse and the dicrotic notch of the PPG pulse if no peak of the APG pulse is near zero is detected.
In another embodiment of the present invention, adjustments are made to the fiducial points based on their positions relative to each other within the PPG pulse waveform. For example, if the diastolic peak of the PPG pulse or dicrotic notch of the PPG pulse occurs before the systolic peak of the PPG pulse, fiducial points on the ascending of the PPG pulse and descending sections of the PPG pulse may be swapped to ensure proper alignment and interpretation.
In another embodiment of the present invention, swapping one or more fiducial points on the ascending section of the PPG pulse with one or more fiducial points on the descending section of the PPG pulse may be implemented as part of the validation process when certain fiducial points are detected in unexpected positions relative to each other within the PPG pulse. In another embodiment of the present invention, if the diastolic peak of the PPG pulse is detected to occur before the systolic peak of the PPG pulse, this indicates a potential anomaly in the waveform. In such cases, swapping fiducial points between the ascending section of the PPG pulse and the descending sections of the PPG pulse can help correct the interpretation and alignment of these points.
In another embodiment of the present invention, swapping one or more fiducial points on the ascending section of the PPG pulse with one or more fiducial points on the descending section of the PPG pulse may be necessary if the dicrotic notch of the PPG pulse is detected to occur before the systolic peak of the PPG pulse within the PPG pulse. In another embodiment of the present invention, the dicrotic notch is a characteristic feature of the PPG waveform that typically follows the systolic peak. If the dicrotic notch is detected prematurely, occurring before the expected systolic peak, this may indicate an anomaly or misalignment in the waveform due to noise.
Further in another embodiment of the present invention, swapping the dicrotic notch of the PPG pulse with the diastolic peak of the PPG pulse may be implemented as a validation step within the fiducial point detection process. This swap occurs if the detected dicrotic notch of the PPG pulse is found to occur after the diastolic peak of the PPG pulse in the waveform. The rationale behind this swap is to ensure the correct alignment and interpretation of fiducial points within the PPG waveform. The dicrotic notch may represent a specific feature of the pulse waveform, occurring after the diastolic peak in normal physiological conditions. If, due to signal variability or other factors, the dicrotic notch is detected in a position that logically corresponds to the diastolic peak, swapping these two points helps to correct any misinterpretation. It ensures that each fiducial point accurately represents its intended physiological characteristic.
FIG. 3E illustrates the feature extraction unit (308) of the application server (104), in accordance with an embodiment of the present subject matter. configured for extracting various features from the validated PPG pulse, each contributing valuable information about cardiovascular dynamics and user-specific characteristics. Further, the feature extraction unit (308) may include extraction of demographic detail (324) and PPG pulse length, Amplitude analysis and Time domain analysis (325). Further, these one or more features may include the rising time and falling time of the PPG pulse, demographic details of the user (such as age and gender), pulse length, and amplitude measurements of key peaks including the systolic peak, diastolic peak, dicrotic notch, and inflection point.
In yet another embodiment of the present invention, one or more features from the PPG pulse may include the rising time of the PPG pulse, falling time of the PPG pulse, demographic details of the user, pulse length of the PPG pulse, amplitude of the systolic peak, amplitude of the diastolic peak, amplitude of the dicrotic notch, amplitude of the inflection point, time of the systolic peak, time of the diastolic peak, time of the dicrotic notch, time of the inflection point, inter-beat interval, inter-beat heart rate < 120, absolute time difference between the systolic peak and the end of the PPG pulse, maxima of a VPG pulse, absolute time difference between the start of the VPG pulse and the w-peak, minima of the VPG pulse, a first area between the start of the PPG pulse till the dicrotic notch, a second area between the dicrotic notch of the PPG pulse till the end of the PPG pulse, and ratio of the first area and the second area.
FIG. 3F illustrates the blood pressure estimation unit (309) of the application server (104), in accordance with an embodiment of the present subject matter is disclosed. The blood pressure estimation unit (309) may be configured to Train/Validate Systolic model using a Random Forest Algorithm (326) and a Train/Validate Diastolic model using the Random Forest Algorithm (327). In one embodiment of the present invention, a random forest algorithm may be used for training the systolic regression model and the diastolic regression model. Further, the one or more features may be applied, blood pressure estimation unit (309), to one of a systolic regression model, a diastolic regression model, and a combination thereof, to estimate the blood pressure of the user. The machine learning models (326, 327) are trained and validated using PPG pulse features extracted from the user, including rising time, falling time, demographic details, pulse length, and various peak amplitudes (systolic, diastolic, dicrotic notch, inflection point), among others. The extracted features are applied to the trained systolic and diastolic regression models, either individually or in combination, to estimate the blood pressure of the user. The models analyze the relationship between the input features and blood pressure values, learning from a labelled dataset to make accurate predictions in real-time scenarios. The machine learning-based blood pressure estimation system may operate in conjunction with smartphones, or other PPG measurement devices, offering convenient and continuous monitoring capabilities. By harnessing the power of machine learning, this approach enables personalized and precise blood pressure assessment without invasive procedures. The machine learning-based blood pressure estimation system may operate in conjunction with smartphones, or other PPG measurement devices, offering convenient and continuous monitoring capabilities. By harnessing the power of machine learning, this approach enables personalized and precise blood pressure assessment without invasive procedures.
In yet another embodiment of the present invention, the method may capture video using integrated one or more optical sensors (202) and flash/LED (205) within a smartphone (203), the described approach may involve utilizing one or more optical sensors (202) and flash/LED (205) components housed within the smartphone (203) for video capture. The one or more optical sensors (202) detect changes in light intensity, particularly from the user's fingertip, while the flash/LED (205) provides illumination for optimal video recording. By integrating these components directly into the smartphone (203), the method ensures efficient and reliable video data collection without the need for external accessories.
Further, the application server (104), via the blood pressure estimation, provides real-time notifications to one or more first users. The real-time notifications are indicative of the status/value of the first information. Further, the application server (104) is configured to receive feedback from the user’s fingertip.
A person skilled in the art will understand that the scope of the disclosure should not be limited to the blood pressure domain and using the aforementioned techniques. Further, the examples provided in supra are for illustrative purposes and should not be construed to limit the scope of the disclosure.
FIGS. 4A-4B is a flowchart that illustrates a method (400) for blood pressure estimation from video photoplethysmography (PPG) using one or more machine learning techniques, in accordance with at least one embodiment of the present subject matter. The flowchart is described in conjunction with FIG. 1 and FIG.3. The method (400) starts at step (401) and proceeds up to step (415).
In operation, the method (400) may involve a variety of steps for blood pressure estimation from video photoplethysmography (PPG) using one or more machine learning models. The method (400) may be implemented by an electronic device (108) including one or more processors (301) and a memory (302) communicatively coupled to the processor (301) and the memory (302) is configured to store processor-executable programmed instructions.
At step (401), the method (400) comprises a step for receiving a video of the user’s fingertip (201). The video is captured using one or more optical sensors (202) coupled with a portable device (203) along with a light source (205).
At step (402), the method (400) comprises a step for extracting a red channel from each frame of one or more frames of the received video.
At step (403), method (400) comprises a step for obtaining a PPG waveform by taking an average of one or more red channels from one or more frames of the received video.
At step (404), the method (400) comprises a step for inverting the PPG waveform to correct inversion format of the PPG waveform and obtain an inverted PPG waveform.
At step (405), the method (400) comprises a step for filtering the inverted PPG waveform to remove noise from the inverted PPG waveform and obtain a filtered PPG waveform.
At step (406), the method (400) comprises a step for selecting the filtered PPG waveform based on detecting signal strength and signal quality.
At step (407), the method (400) comprises a step for extracting a portion of the PPG pulse from the selected PPG waveform.
At step (408), the method (400) comprises a step for performing a pulse correction on the PPG pulse. The pulse correction is performed by the following steps.
At step (409), the method (400) comprises a step for detecting peaks and troughs from the PPG pulse. A peak is the highest point between two troughs.
At step (410), the method (400) comprising a step for detecting an ascending section and a descending section of the PPG pulse, wherein the ascending section is starting from the first trough of the PPG pulse ending at the peak of the PPG pulse and the descending section is starting from the peak of the PPG pulse ending at a second trough of the PPG pulse.
At step (411), the method (400) comprises a step for calculating area of the ascending section and the descending section of the PPG pulse.
At step (412), the method (400) comprises a step for flipping the PPG pulse if area of the ascending section of the PPG pulse is greater than the area of the descending section of the PPG pulse.
At step (413), the method (400) comprises a step for validating the PPG pulse through fiducial point detection, and fiducial point validation by defining thresholds on the fiducial points.
At step (414), the method (400) comprises a step for extracting one or more features from the validated PPG pulse.
At step (415), the method (400) comprises a step for applying one or more extracted features to one of a systolic regression model, a diastolic regression model, and a combination thereof, to estimate the blood pressure of the user.
Let us delve into a detailed working example of the present disclosure.
Example: X is a person with hypertension or other cardiovascular conditions who needs to monitor their blood pressure levels regularly. X places the fingertip on the optical sensor of the portable device and initiates the video capture process. The device's optical sensor captures a high-resolution video of Sarah's fingertip, illuminated by the built-in light source. The recorded video is processed to extract the red channel information, which primarily represents the changes in blood volume in X's fingertip over time. The extracted red channel data is analysed to generate a PPG waveform, reflecting the pulsatile changes in blood volume associated with X's heartbeat. In some cases, the PPG waveform may be inverted due to technical factors. The method corrects any inversion in the waveform to ensure an accurate interpretation of the blood volume changes. To enhance the quality of the PPG waveform, noise present in the signal, such as motion artefacts or ambient light interference, is filtered out. The method employs algorithms to select the most reliable segment of the PPG waveform for further analysis. Factors like signal strength and signal-to-noise ratio are considered during this selection process. From the selected portion of the PPG waveform, the distinct pulsatile component corresponding to X's heartbeat is isolated. The detected pulse is further refined to ensure accurate identification of peaks and troughs, essential for precise blood pressure estimation. Fiducial points, such as the systolic peak and diastolic trough, are detected within the PPG pulse. These points are then validated against predefined criteria to ensure their accuracy. Various features are extracted from the validated PPG pulse, including pulse amplitude, duration, and shape characteristics. Utilising machine learning models trained on a dataset of PPG features and corresponding blood pressure measurements, X's systolic and diastolic blood pressure levels are estimated.
Finally, armed with these features, the system employs a machine learning model to estimate X's blood pressure levels based on the information gathered from the video and pulse analysis.
A person skilled in the art will understand that the scope of the disclosure is not limited to scenarios based on the aforementioned factors and using the aforementioned techniques and that the examples provided do not limit the scope of the disclosure.
FIG. 5 illustrates a block diagram of an exemplary computer system (501) for implementing embodiments consistent with the present disclosure.
Variations of computer systems (501) may be used for securely performing one or more operations in a user interface platform. The computer system (501) may comprise a central processing unit (“CPU” or “processor”) (502). The processor (502) may comprise at least one data processor for executing program components for executing user- or system-generated requests. A user may include a person, a person using a device such as such as those included in this disclosure, or such a device itself. Additionally, the processor (502) may include specialized processing units such as integrated system (bus) controllers, memory management control units, floating point units, graphics processing units, digital signal processing units, or the like. In various implementations, the processor (502) may include a microprocessor, such as AMD Athlon, Duron or Opteron, ARM’s application, embedded or secure processors, IBM PowerPC, Intel’s Core, Itanium, Xeon, Celeron or other line of processors, for example. Accordingly, the processor (502) may be implemented using a mainframe, distributed processor, multi-core, parallel, grid, or other architectures. Some embodiments may utilize embedded technologies like application-specific integrated circuits (ASICs), digital signal processors (DSPs), or Field Programmable Gate Arrays (FPGAs), for example.
Processor (502) may be disposed of in communication with one or more input/output (I/O) devices via an I/O interface (503). Accordingly, the I/O interface (503) may employ communication protocols/methods such as, without limitation, audio, analog, digital, monoaural, RCA, stereo, IEEE-1394, serial bus, universal serial bus (USB), infrared, PS/2, BNC, coaxial, component, composite, digital visual interface (DVI), high-definition multimedia interface (HDMI), RF antennas, S-Video, VGA, IEEE 802.n /b/g/n/x, Bluetooth, cellular (e.g., code-division multiple access (CDMA), high-speed packet access (HSPA+), global system for mobile communications (GSM), long-term evolution (LTE), WiMAX, or the like, for example.
Using the I/O interface (503), the computer system (501) may communicate with one or more I/O devices. For example, the input device (504) may be an antenna, keyboard, mouse, joystick, (infrared) remote control, camera, card reader, fax machine, dongle, biometric reader, microphone, touch screen, touchpad, trackball, sensor (e.g., accelerometer, light sensor, GPS, gyroscope, proximity sensor, or the like), stylus, scanner, storage device, transceiver, video device/source, or visors, for example. Likewise, an output device (505) may be a user’s smartphone, tablet, cell phone, laptop, printer, fax machine, video display (e.g., cathode ray tube (CRT), liquid crystal display (LCD), light-emitting diode (LED), plasma, or the like), or audio speaker, for example. In some embodiments, a transceiver (506) may be disposed of in connection with the processor (502). The transceiver (506) may facilitate various types of wireless transmission or reception. For example, the transceiver (506) may include an antenna operatively connected to a transceiver chip (example devices include the Texas Instruments® WiLink WL1283, Broadcom® BCM4750IUB8, Infineon Technologies® X-Gold 618-PMB9800, or the like), providing IEEE 802.11a/b/g/n, Bluetooth, FM, global positioning system (GPS), and/or 2G/3G/5G/6G HSDPA/HSUPA communications, for example.
In some embodiments, the processor (502) may be disposed of in communication with a communication network (508) via a network interface (507). The network interface (507) is adapted to communicate with the communication network (508). The network interface (507) may employ connection protocols including, without limitation, direct connect, Ethernet (e.g., twisted pair 10/100/1000 Base T), transmission control protocol/internet protocol (TCP/IP), token ring, or IEEE 802.11a/b/g/n/x, for example. The communication network (508) may include, without limitation, a direct interconnection, local area network (LAN), wide area network (WAN), wireless network (e.g., using Wireless Application Protocol), or the Internet, for example. Using the network interface (507) and the communication network (508), the computer system (501) may communicate with devices such as shown as a laptop (509) or a mobile/cellular phone (510). Other exemplary devices may include, without limitation, personal computer(s), server(s), fax machines, printers, scanners, various mobile devices such as cellular telephones, smartphones (e.g., Apple iPhone, Blackberry, Android-based phones, etc.), tablet computers, eBook readers (Amazon Kindle, Nook, etc.), laptop computers, notebooks, gaming consoles (Microsoft Xbox, Nintendo DS, Sony PlayStation, etc.), or the like. In some embodiments, the computer system (501) may itself embody one or more of these devices.
In some embodiments, the processor (502) may be disposed of in communication with one or more memory devices (e.g., RAM 413, ROM 414, etc.) via a storage interface (512). The storage interface (512) may connect to memory devices including, without limitation, memory drives, removable disc drives, etc., employing connection protocols such as serial advanced technology attachment (SATA), integrated drive electronics (IDE), IEEE-1394, universal serial bus (USB), fiber channel, small computer systems interface (SCSI), etc. The memory drives may further include a drum, magnetic disc drive, magneto-optical drive, optical drive, redundant array of independent discs (RAID), solid-state memory devices, or solid-state drives, for example.
The memory devices may store a collection of program or database components, including, without limitation, an operating system (516), user interface application (517), web browser (518), mail client/server (519), user/application data (520) (e.g., any data variables or data records discussed in this disclosure) for example. The operating system (516) may facilitate resource management and operation of the computer system (501). Examples of operating systems include, without limitation, Apple Macintosh OS X, UNIX, Unix-like system distributions (e.g., Berkeley Software Distribution (BSD), FreeBSD, NetBSD, OpenBSD, etc.), Linux distributions (e.g., Red Hat, Ubuntu, Kubuntu, etc.), IBM OS/2, Microsoft Windows (XP, Vista/7/8, etc.), Apple iOS, Google Android, Blackberry OS, or the like.
The user interface (517) is for facilitating the display, execution, interaction, manipulation, or operation of program components through textual or graphical facilities. For example, user interfaces may provide computer interaction interface elements on a display system operatively connected to the computer system (501), such as cursors, icons, check boxes, menus, scrollers, windows, or widgets, for example. Graphical user interfaces (GUIs) may be employed, including, without limitation, Apple Macintosh operating systems’ Aqua, IBM OS/2, Microsoft Windows (e.g., Aero, Metro, etc.), Unix X-Windows, or web interface libraries (e.g., ActiveX, Java, JavaScript, AJAX, HTML, Adobe Flash, etc.), for example.
In some embodiments, the computer system (501) may implement a web browser (518) stored program component. The web browser (518) may be a hypertext viewing application, such as Microsoft Internet Explorer, Google Chrome, Mozilla Firefox, Apple Safari, or Microsoft Edge, for example. Secure web browsing may be provided using HTTPS (secure hypertext transport protocol), secure sockets layer (SSL), Transport Layer Security (TLS), or the like. Web browsers may utilize facilities such as AJAX, DHTML, Adobe Flash, JavaScript, Java, or application programming interfaces (APIs), for example. In some embodiments, the computer system (501) may implement a mail client/server (519) stored program component. The mail server (519) may be an Internet mail server such as Microsoft Exchange, or the like. The mail server may utilize facilities such as ASP, ActiveX, ANSI C++/C#, Microsoft .NET, CGI scripts, Java, JavaScript, PERL, PHP, Python, or WebObjects, for example. The mail server (519) may utilize communication protocols such as internet message access protocol (IMAP), messaging application programming interface (MAPI), Microsoft Exchange, post office protocol (POP), simple mail transfer protocol (SMTP), or the like. In some embodiments, the computer system (501) may implement a mail client (520) stored program component. The mail client (520) may be a mail viewing application, such as Apple Mail, Microsoft Entourage, Microsoft Outlook, or Mozilla Thunderbird.
In some embodiments, the computer system (501) may store user/application data (521), such as the data, variables, records, or the like as described in this disclosure. Such databases may be implemented as fault-tolerant, relational, scalable, secure databases such as Oracle or Sybase, for example. Alternatively, such databases may be implemented using standardized data structures, such as an array, hash, linked list, struct, structured text file (e.g., XML), table, or as object-oriented databases (e.g., using ObjectStore, Poet, Zope, etc.). Such databases may be consolidated or distributed, sometimes among the various computer systems discussed above in this disclosure. It is to be understood that the structure and operation of any computer or database component may be combined, consolidated, or distributed in any working combination.
Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present invention. A computer-readable storage medium refers to any type of physical memory on which information or data readable by a processor may be stored. Thus, a computer-readable storage medium may store instructions for execution by one or more processors, including instructions for causing the processor(s) to perform steps or stages consistent with the embodiments described herein. The term “computer- readable medium” should be understood to include tangible items and exclude carrier waves and transient signals, i.e., non-transitory. Examples include Random Access Memory (RAM), Read- Only Memory (ROM), volatile memory, nonvolatile memory, hard drives, Compact Disc (CD) ROMs, Digital Video Disc (DVDs), flash drives, disks, and any other known physical storage media.
Various embodiments of the disclosure encompass numerous advantages including methods and systems for blood pressure estimation using photoplethysmography (PPG). The disclosed method and system have several technical advantages, but not limited to the following:
• Unlike traditional blood pressure measurement techniques that require cuff-based devices, the disclosed invention provides a non-invasive approach using video photoplethysmography (PPG). This eliminates discomfort and potential risks associated with cuff-based measurements.
• By utilizing a smartphone's camera and optical sensors, the invention enables easy and convenient blood pressure monitoring anytime and anywhere. Users can simply use their smartphone to capture PPG signals without the need for additional specialized equipment.
• The invention enhances accessibility to blood pressure monitoring, especially in remote or underserved areas where access to medical facilities may be limited. It empowers individuals to monitor their blood pressure conveniently using widely available technology.
• Leveraging machine learning models with video PPG signals, the invention allows for continuous and real-time blood pressure estimation. This real-time monitoring capability is valuable for tracking changes in blood pressure and detecting abnormalities promptly.
• The invention integrates seamlessly with smartphones and other smart devices, enabling integration with existing digital health platforms and applications. This facilitates personalized health management and data sharing for healthcare professionals.
• By harnessing advancements in video PPG and machine learning, the invention represents an innovative approach to blood pressure estimation. It showcases the potential of emerging technologies to improve healthcare outcomes and patient engagement.
Various modifications to the embodiment will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. However, one of ordinary skill in the art will readily recognize that the present disclosure is not intended to be limited to the embodiments illustrated but is to be accorded the widest scope consistent with the principles and features described herein.
The foregoing description shall be interpreted as illustrative and not in any limiting sense. A person of ordinary skill in the art would understand that certain modifications could come within the scope of this disclosure.
The embodiments, examples, and alternatives of the preceding paragraphs or the description, including any of their various aspects or respective individual feature(s), may be taken independently or in any combination. Features described in connection with one embodiment are applicable to all embodiments unless such features are incompatible.
, Claims:WE CLAIM
1. A method (400) for blood pressure estimation from video photoplethysmography (PPG) using one or more machine learning techniques, the method (400) comprises:
receiving (401) by a processor (301), a video of user's fingertip (201), wherein the video is captured using one or more optical sensors (202) coupled with a portable device (203) along with a light source (205);
extracting (402), by the processor (301), a red channel from each frame of one or more frames of the received video;
obtaining (403), by the processor (301), a PPG waveform by taking average of one or more red channels from the one or more frames of the received video;
inverting (404), by the processor (301), the PPG waveform to correct inversion format of the PPG waveform and obtain an inverted PPG waveform;
filtering (405), by the processor (301), the inverted PPG waveform to remove noise from the inverted PPG waveform and obtain a filtered PPG waveform;
selecting (406), by the processor (301), the filtered PPG waveform based on detecting signal strength and signal quality;
extracting (407), by the processor (301), a portion of PPG pulse from the selected PPG waveform;
performing (408), by the processor (301), pulse correction on the PPG pulse, wherein the pulse correction is performed by:
detecting (409), by the processor (301), peaks and troughs from the PPG pulse, wherein a peak is the highest point between two troughs;
detecting (410), by the processor (301), an ascending section and a descending section of the PPG pulse, wherein the ascending section is starting from a first trough of the PPG pulse ending at the peak of the PPG pulse and the descending section is starting from the peak of the PPG pulse ending at a second trough of the PPG pulse;
calculating (411), by the processor (301), areas of both the ascending section and the descending section of the PPG pulse;
flipping (412), by the processor (301), the PPG pulse if area of the ascending section is greater than the area of the descending section;
validating (413), by the processor (301), the PPG pulse through fiducial point detection, fiducial point validation by defining thresholds on the fiducial points;
extracting (414), by the processor (301), one or more features from the validated PPG pulse; and
applying (415), by the processor (301), the extracted one or more features to one of a systolic regression model, a diastolic regression model, and a combination thereof, to estimate the blood pressure of the user.
2. The method (400) as claimed in claim 1, wherein filtering the inverted PPG waveform is performed using on one or more filtering techniques; wherein one or more filtering techniques correspond to one of bandpass filtering, signal smoothing, signal interpolation and a combination thereof.
3. The method (400) as claimed in claim 1, comprises detecting the signal strength of the filtered PPG waveform by quantifying noise present in the filtered PPG waveform; wherein the method (400) comprises determining signal quality of the filtered PPG waveform by applying techniques selected from one of thresholding, kurtosis, entropy, baseline wander removal and a combination thereof.
4. The method (400) as claimed in claim 1, wherein extracting the portion of PPG pulse from the selected PPG waveform is performed by sliding a window of a predetermined time-period over the selected PPG waveform without overlapping.
5. The method (400) as claimed in claim 1, wherein the ascending section corresponds to start of the PPG pulse to the peak of the PPG pulse; wherein the descending section corresponds to peak of the PPG pulse to an end of the PPG pulse.
6. The method (400) as claimed in claim 1,
wherein the fiducial points of the PPG pulse comprise start point of the PPG pulse, maximum slope point, systolic peak, inflection point, dicrotic notch, diastolic peak, and end point of the PPG pulse;
wherein fiducial points of a velocity plethysmogram (VPG) pulse comprises u-peak, v-peak, and w-peak;
wherein fiducial points of an acceleration plethysmogram (APG) pulse comprises a-peak, b-peak, c-peak, d-peak, and e-peak;
wherein the VPG pulse is a first derivative of the PPG pulse and the APG pulse is a second derivative of the PPG pulse.
7. The method (400) as claimed in claim 6, wherein detection of the fiducial points of the PPG pulse is performed by:
finding the start point and the end point of the PPG pulse;
finding the maximum slope point of the PPG pulse based on maxima of the VPG pulse, wherein the maxima of the VPG pulse corresponds to u-peak of the VPG pulse;
finding the diastolic peak of the PPG pulse based on minima of the APG pulse, wherein the minima of the APG pulse corresponds to e-peak of the APG pulse;
finding the dicrotic notch of the PPG pulse based on coinciding minima point of the VPG pulse with the PPG pulse, wherein the minima point of the VPG pulse corresponds to the v-peak of the VPG pulse; and
finding the inflection point of the PPG pulse based on a peak of the APG pulse close to a value zero, wherein the peak of the APG pulse close to zero corresponds to one of the c-peak, d-peak or a combination thereof.
8. The method (400) as claimed in claim 7, wherein validation of the fiducial points through defining thresholds on the fiducial points of the PPG pulse is performed by:
considering extrema (the last minima) of the APG pulse as the diastolic peak of the PPG pulse, if more than one minima are available on the APG pulse;
ensuring the diastolic peak of the PPG pulse to be not equal to either the maximum slope point of the PPG pulse or the systolic peak of the PPG pulse;
considering local minima of either the APG pulse or the VPG pulse, as the dicrotic notch of the PPG pulse, if the minima point of the VPG pulse does not coincide with the PPG pulse;
ensuring the dicrotic notch of the PPG pulse to be lies between the systolic peak and the diastolic peak of the PPG pulse;
considering an average of the diastolic peak of the PPG pulse and the dicrotic notch of the PPG pulse, as the inflection point of the PPG pulse, if no peak of the APG pulse close to the value zero;
swapping one or more fiducial points on the ascending section of the PPG pulse with one or more fiducial points on the descending section of the PPG pulse, if the diastolic peak of the PPG pulse lies before the systolic peak of the PPG pulse;
swapping one or more fiducial points on the ascending section of the PPG pulse with one or more fiducial points on the descending section of the PPG pulse, if dicrotic notch of the PPG pulse lies before the systolic peak of the PPG pulse; and
swapping dicrotic notch of the PPG pulse with diastolic peak of the PPG pulse, if dicrotic notch of the PPG pulse lies after the diastolic peak of the PPG pulse.
9. The method (400) as claimed in claim 1, wherein one or more features from the PPG pulse comprise rising time of the PPG pulse, falling time of the PPG pulse, demographic details of the user, pulse length of the PPG pulse, amplitude of the systolic peak, amplitude of the diastolic peak, amplitude of the dicrotic notch, amplitude of the inflection point, time of the systolic peak, time of the diastolic peak, time of the dicrotic notch, time of the inflection point, inter-beat interval, inter-beat heart rate < 120, absolute time difference between the systolic peak and the end of the PPG pulse, maxima of a VPG pulse, absolute time difference between the start of the VPG pulse and the w-peak, minima of the VPG pulse, a first area between the start of the PPG pulse till the dicrotic notch, a second area between the dicrotic notch of the PPG pulse till the end of the PPG pulse, and ratio of the first area and the second area.
10. The method (400) as claimed in claim 1, comprises training the systolic regression model and the diastolic regression model, using the one or more machine learning techniques; wherein the one or more machine learning technique correspond to random forest algorithm.
11. The method (400) as claimed in claim 1, comprises capturing the video using the one or more optical sensors (202) and the light source (205) placed within the portable device (203); wherein the portable device (203) corresponds to a smartphone, wherein the one or more optical sensors (202) corresponds to rear camera of a smartphone, wherein the light source (205) corresponds to a flashlight placed on rear side of the smartphone.
12. The method (400) as claimed in claim 1, wherein inverting (404) the PPG waveform corresponds to vertically inverting the PPG waveform; wherein inverting (404) the PPG waveform corresponds to inverting polarity of the PPG waveform.
13. The method (400) as claimed in claim 1, wherein flipping (412) of the PPG pulse corresponds to horizontally flipping the PPG pulse; wherein flipping (412) of the PPG pulse corresponds to reversing phase of the PPG pulse.
14. A system (100) for blood pressure estimation from video photoplethysmography (PPG) using one or more machine learning techniques, the system (100) comprises:
one or more portable devices (203), wherein one or more portable devices (203) are coupled with one or more optical sensors (202) and a light source (205);
a memory (302);
a processor (301), wherein the processor (301) is configured to execute programmed instructions stored in the memory (302), by:
receiving (401) a video of user's fingertip (201), wherein the video is captured using the one or more optical sensors (202) coupled with the one or more portable devices (203) along with the light source (203);
extracting (402) a red channel from each frame of one or more frames of the received video;
obtaining (403) a PPG waveform by taking average of one or more red channels from one or more frames of the received video;
inverting (404) the PPG waveform to correct inversion format of the PPG waveform and obtain an inverted PPG waveform;
filtering (405) the inverted PPG waveform to remove noise from the inverted PPG waveform and obtain a filtered PPG waveform;
selecting (406) the filtered PPG waveform based on detecting signal strength and signal quality;
extracting (407) a portion of PPG pulse from the selected PPG waveform;
performing (408) pulse correction on the PPG pulse, wherein the pulse correction is performed by:
detecting (409) peaks and troughs from the PPG pulse, wherein a peak is the highest point between two troughs;
detecting (410) an ascending section and a descending section of the PPG pulse, wherein the ascending section is starting from a first trough of the PPG pulse ending at the peak of the PPG pulse and the descending section is starting from the peak of the PPG pulse ending at a second trough of the PPG pulse;
calculating (411) areas of both the ascending section and the descending section of the PPG pulse;
flipping (412) the PPG pulse if area of the ascending section is greater than the area of the descending section;
validating (413) the PPG pulse through fiducial point detection, fiducial point validation by defining thresholds on the fiducial points;
extracting (414) one or more features from the validated PPG pulse; and
applying (415) the extracted one or more features to one of a systolic regression model, a diastolic regression model, and a combination thereof, to estimate the blood pressure of the user.

Dated this 11th day of July 2024

Abhijeet Gidde
Agent for the Applicant
IN/PA-4407