DESC:EARLIEST PRIORITY DATE:
This Application claims priority from a Provisional patent application filed in India having Patent Application No. 202141047953, filed on October 21, 2021 and titled “SYSTEM AND METHOD FOR CARDIAC DISEASE AND ASTHMA PREDICTIONS”
FIELD OF INVENTION
[0001] Embodiments of a present disclosure relate to devices for monitoring health of human beings, and more particularly to a device for monitoring and prognosis of health and method to operate the same.
BACKGROUND
[0002] Health is a state of complete mental, physical, and social well-being. Physical health in human includes physical activity, nutrition and diet, medical self-care, rest, and sleep. Mental health is a level of psychological well-being. Health problems refers to health issues related to human anatomy.
[0003] Several heath issues exist all over the world. Most common health issue is ‘disease’. A disease may be contagious or non-contagious. A vast number of people die each year due to non-contagious diseases. Many diseases are related to mental and physical health. The diseases like cardiac diseases, diabetes and asthma are also related to stress. Approximately, 4 lakhs people in the world dies due to cardiac and asthma attack. Although, this death rate is maximum in people above 60 years, the youth of today’s era are also suffering from cardiac and asthmatic problems due to lifestyle disorder. The problem increases when, the health is not monitored on regular basis.
[0004] There are many devices used for monitoring cardiac movement like, ECG (Electrocardiogram), stress test, Echocardiogram etc. Also, there are many techniques are available to monitor asthma patients, such as FeNO, spirometry and the like. In FeNO, a patient breathes into a machine that measures the level of nitric oxide in patient’s breathe, which is a sign of inflammation in his/her lungs. Another technique is spirometry, in which patient blow into a machine that measures how fast the patient may breathe out and how much air the patient may hold in his/her lungs.
[0005] Although, the currently used health monitoring devices are useful for monitoring patient’s health. However, there are various limitations to the currently existing health monitoring devices. For using these devices and methods, a patient must visit the hospital physically. In cases where patient is not able to go in the hospital, it is a tough task to monitor patients’ cardiac, asthmatic, and overall health. If a device is bought to home for monitoring health, a medical expert is needed for operating the device. Accordingly, the currently existing devices are inefficient to prognosis health problems that may occur in the future.
[0006] Hence, there is a need for an improved device and method for monitoring and predicting health problems which addresses the aforementioned issues.
BRIEF DESCRIPTION
[0007] In accordance with one embodiment of the disclosure, a device for monitoring and prognosis of health is provided. The device includes a housing unit and a processor unit. The housing unit includes an interface unit, a filtration unit, a strain sensor, and a converter unit. The interface unit is configured to obtain a plurality of auscultation signals of the user. The filtration unit is operatively coupled to the interface unit. The filtration unit is configured to eliminate noise signals from the plurality of auscultation signals. The strain sensor is configured to sense one or more physiological signals corresponding to stress level of the user by measuring skin resistance variations. The converter unit is operatively coupled with the filtration unit and strain sensor. The converter unit is configured to digitalize the plurality of auscultation signals filtered by the filtration unit and the one or more physiological signals sensed by the strain sensor. The processor unit includes an integration module, a comparator module, and a prediction module. The integration module is configured to integrate the plurality of auscultation signals, and the one or more physiological signals digitized by the converter unit to obtain an integrated health signal. The comparator module is configured to compare the integrated health signals with historical health parameter datasets using a pre-trained model to obtain a comparative health dataset. The pre-trained model is trained with a plurality of health-issue-related datasets using a convolution neural network (CNN) technique in an iterative manner. The prediction module is configured to predict the one or more health issues corresponding to the user based on the comparative health dataset, thereby performing monitoring and prognosis of health of the user. The one or more health issues is corresponding to at least one of heart dysfunction, asthma, pneumonia, and stress. of the user.
[0008] In accordance with another embodiment, a method for monitoring and prognosis of health is provided. The method includes obtaining, by an interface unit, a plurality of auscultation signals of health parameters of a use. The method also includes, filtering, by a filtration unit, to eliminate noise signals from the plurality of auscultation signals obtained by the interface unit. Further, the method also includes, sensing, by a strain sensor, to sense one or more physiological signals corresponding to stress level of the user by measuring skin resistance variations. Furthermore, the method also includes, converting, by a converter unit, to digitalise the plurality of auscultation signals filtered by filtration unit and the one or more physiological signals by strain sensor. The method includes, integrating, by an integration module, to integrate the plurality of auscultation signals and the one or more physiological signals digitized by the converter unit to obtain an integrated health signal. Moreover, the method also includes, comparing, by a comparator module, to compare the integrated health signals with historical health parameter datasets using a pre-trained model to obtain a comparative health dataset where the pre-trained model is trained with a plurality of health-issue-related datasets using a convolution neural network (CNN) technique in an iterative manner. Furthermore, the method also includes, predicting, by a prediction module, to predict one or more health issues corresponding to the user based on the comparative health dataset, thereby performing monitoring and prognosis of health of the user where the one or more health issues is corresponding to at least one of heart dysfunction, asthma, pneumonia, and stress of the user.
[0009] To further clarify the advantages and features of the present disclosure, a more particular description of the disclosure will follow by reference to specific embodiments thereof, which are illustrated in the appended figures. It is to be appreciated that these figures depict only typical embodiments of the disclosure and are therefore not to be considered limiting in scope. The disclosure will be described and explained with additional specificity and detail with the appended figures.
BRIEF DESCRIPTION OF THE DRAWINGS
[0010] The disclosure will be described and explained with additional specificity and detail with the accompanying figures in which:
[0011] FIG. 1 is a block diagram representing a device for monitoring and prognosis of health in accordance with an embodiment of the present disclosure;
[0012] FIG. 2 is a block diagram representing another embodiment of the device for monitoring and prognosis of health of FIG. 1 in accordance with an embodiment of the present disclosure;
[0013] FIG. 3 is a block diagram representing an exemplary embodiment of the device for monitoring and prognosis of health of FIG. 1 in accordance with an embodiment of the present disclosure;
[0014] FIG. 4 is a block diagram of a computer or a server for the device for monitoring and prognosis of health in accordance with an embodiment of the present disclosure; and
[0015] FIG. 5 is a flow chart representing steps involved in a method for operation of the device for monitoring and prognosis of health.
[0016] Further, those skilled in the art will appreciate that elements in the figures are illustrated for simplicity and may not have necessarily been drawn to scale. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the figures by conventional symbols, and the figures may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the figures with details that will be readily apparent to those skilled in the art having the benefit of the description herein.
DETAILED DESCRIPTION
[0017] For the purpose of promoting an understanding of the principles of the disclosure, reference will now be made to the embodiment illustrated in the figures and specific language will be used to describe them. It will nevertheless be understood that no limitation of the scope of the disclosure is thereby intended. Such alterations and further modifications in the illustrated system, and such further applications of the principles of the disclosure as would normally occur to those skilled in the art are to be construed as being within the scope of the present disclosure.
[0018] The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such a process or method. Similarly, one or more devices or sub-systems or elements or structures or components preceded by "comprises... a" does not, without more constraints, preclude the existence of other devices, sub-systems, elements, structures, components, additional devices, additional sub-systems, additional elements, additional structures, or additional components. Appearances of the phrase "in an embodiment", "in another embodiment" and similar language throughout this specification may, but not necessarily do, all refer to the same embodiment.
[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this disclosure belongs. The system, methods, and examples provided herein are only illustrative and not intended to be limiting.
[0020] In the following specification and the claims, reference will be made to a number of terms, which shall be defined to have the following meanings. The singular forms “a”, “an”, and “the” include plural references unless the context clearly dictates otherwise.
[0021] Embodiments of the present disclosure relate to a device for monitoring and prognosis of health and method to operate the same. As used herein, the term “health monitoring and prognosis” denotes observation and predictive diagnosis of current state of health of a user. Further, the device is described hereafter in FIG. 1 is a device for monitoring and prognosis of health and method to operate the same.
[0022] FIG. 1 is a block diagram representing a device (100) for monitoring and prognosis of health in accordance with an embodiment of the present disclosure. The device (100) includes a housing unit (102) and a processor unit (112).
[0023] The housing unit (102) includes an interface unit (104), a filtration unit (106), a strain sensor (108) and a converter unit (110). The interface unit (104) is configured to obtain a plurality of auscultation signals of a user. The plurality of auscultation signals are sound signals from a heart, lungs of the user. When the device (100) is in an operative position, the interface unit (104) obtains the plurality of auscultation signals such as sound of heart beats and sound of breathing from the user.
[0024] Such sound signals contain noise signals which is obtained at the time of obtaining the plurality of auscultation signals. The filtration unit (106) is operatively coupled to the interface unit (104). The filtration unit (106) is configured to eliminate noise signals from the plurality of auscultation signals. Further, the strain sensor (108) is configured to sense one or more physiological signals corresponding to stress level of the user by measuring skin resistance variations. The body has resistance to current flow. More than 99% of the body's resistance to electric current flow is at the skin. The resistance of the body, including the resistance of the skin, determines to a major degree the amount of the current which may flow through the human body. The skin resistance varies with the state of sweat glands in the skin. Sweating is controlled by the sympathetic nervous system and skin conductance is an indication of psychological or physiological arousal. If the sympathetic branch of the autonomic nervous system is highly aroused, then sweat gland activity also increases, which in turn increases skin conductance. In this way, skin conductance may be a measure of stress in the user.
[0025] In one embodiment, the strain sensor (108) may be stress sensor. The stain sensor (108) senses the vibrations of skin and measures the level of sensed vibrations. The device (100) when in the operative position, the skin variation of the user apply force which is measured by the strain sensor (108). The resistance of strain sensor (108) varies with the force applied on the strain sensor (108). The strain sensor (108) converts the force into electronic signals. The measured level of stress/ strain then sent to the converter unit (110).
[0026] Furthermore, the converter unit (110) is operatively coupled with the filtration unit (106) and the strain sensor (108). The converter unit (110) is configured to digitalize the plurality of auscultation signals filtered by the filtration unit (106) and the one or more physiological signals sensed by the strain sensor (108). In one embodiment, the converter unit (110) take the plurality of auscultation signals as an analogue sound signal and produces a digital sound signal. The number of binary digits, or bits used to represent the analogue signals value depends on the resolution of an analogue to digital converter (ADC) of the converter unit (110). In an exemplary embodiment, a 4-bit ADC may have a resolution whereas an 8-bit ADC may have low resolution. Thus, an analogue to digital converter takes an unknown continuous analogue signal and converts the signals into an “n”- bit binary number of 2n bits. In one embodiment, the converter (110) may be a 24 -bit converter to digitalise the plurality of auscultation signals.
[0027] The processor unit (112) is communicably coupled to the housing unit (102). In one embodiment, the process unit may be located locally within the housing unit (102). In such an embodiment, the processor unit (112) may be a microcontroller. In another embodiment, the processor unit (112) may be in a cloud-based server. The processor unit (112) includes an integration module (114), a comparator module (116), and a prediction (118).
[0028] The integration module (114) is configured to integrate the plurality of auscultation signals and the one or more physiological signals digitized by the converter unit (110) to obtain an integrated health signal. The comparator module (116) is configured to compare the integrated health signals with historical health parameter datasets using a pre-trained model to obtain a comparative health dataset. The pre-trained model is trained with a plurality of health-issue-related datasets using a convolution neural network (CNN) technique in an iterative manner. The training of pre-trained model is done using a low power and low latency CNN based modelling. The trained CNN model is saved and may be compared with input signal (sound signals and signals obtained from strain sensor (108) which is again modelled with the modified CNN. The prediction module (118) is configured to predict one or more health issues corresponding to the user based on the comparative health dataset, thereby performing monitoring and prognosis of health of the user. The one or more health issues is corresponding to at least one of heart dysfunction, asthma, pneumonia, and stress, of the user. The health signals from comparative dataset are processed using low power support vector machine (SVM) and artificial intelligence to predict heart and lung disorder.
[0029] The SVMs are supervised learning models with associated learning process that analyse data for classification and regression analysis. The regression analysis is a statistical process for estimating relationship between the health dataset received and the historical health data stored in the storage unit (128). As used herein, the “health prediction” refers to the output of a processing of health dataset after the dataset may be trained on a historical health dataset and applied to new data when forecasting health. In one exemplary embodiment, an artificial intelligence (AI) platform prediction manages computing resources in the cloud to run the trained models. The predictions may be requested from the trained models and may get predicted target values of health parameters. In one embodiment, the process to make health predictions, includes exporting of trained model as artifacts in the cloud that may be deployed onto the AI platform prediction. The processes also include creating a model resource in AI platform prediction. The modification in the trained model may be saved. The new model may be created from the saved model. For online prediction, the service runs the saved model and returns the requested predictions. Thus, the health prediction may be done using the SVM and the AI platform prediction.
[0030] FIG. 2 is a block diagram representing another embodiment of the device (100) for monitoring and prognosis of health of FIG. 1 in accordance with an embodiment of the present disclosure. The device (100) of FIG. 1 includes housing unit (102) and a processor unit (112). In one embodiment, the device (100) of FIG. 1 may include a charging unit (130) configured to charge a battery (136). In some embodiments, the battery (136) may be a lithium-ion battery. In another embodiment, the device (100) may also include a booster circuit (132) configured to boost power from the charging unit (130) to the processor unit (112).
[0031] In one embodiment, the interface unit (104) may be a diaphragm (138) in a microphone (140). In another embodiment, the interface unit (104) includes a microphone (140). The microphone (140) is adapted to absorb sound from the heart or the lung of the user. As used herein, in the microphone (140), sound waves (sound-pressure variations in the air) are converted into corresponding variations in electric current in two operations which take place almost simultaneously. In the first, the sound wave impinges on the diaphragm (138), causing the diaphragm (138) to move to and from, in a manner corresponding to the movement of the air particles. In the second, the diaphragm (138) by its motion causes a corresponding change in some property of an electric circuit. In each case, motion of the diaphragm (138) produces a variation in the electric output. In one embodiment of the present disclosure, the interface unit (104) is provided with a plurality of modes such as a heart mode and a lung mode. Based on the mode activated, the interface unit (104) absorbs the plurality of auscultation signals from either the lung or the heart and sends the plurality of auscultation signals to the filtration unit (106). However, the present disclosure is not limited to any particular interface unit (104) described.
[0032] In one embodiment, the filtration unit (106) includes an amplifier (120) to amplify the plurality of auscultation signals. The amplifier (120) uses electric power and increase the amplitude of the plurality of auscultation signals, hence increase in power of the signals. In some embodiments, the filtration unit (106) includes a first level filter (106 a) configured to minimize the noise from the plurality of signals received from the interface unit (104). In such an embodiment, the filtration unit (106) may also include a second level filter (106 b) configured to reduce the noise from amplified signals. In one embodiment, the first level filter (106 a) is a low power modified CMOS (complementary metal-oxide semiconductor) programmable band pass filter and second level filter (106 b) is a programmable Butterworth filter. As used herein, the band pass filter allows the plurality of auscultation signals to pass, which are within a certain range and rejects the signals which are outside that range. The amplifier (120) then increases amplitude such as making the plurality of auscultation signals strong. Similarly, as used herein, the Butterworth filter is a type of signal processing filter designed to have a purer audio as an output for the possible range of frequencies which may pass through the filter. The Butterworth filter then filters the amplified signals to obtain pure form of the plurality of auscultation signals.
[0033] In one embodiment, the device (100) includes a display device (124) which is coupled to the processor unit (112). The display device (124) is configured to prediction of the one or more health issues corresponding to the user. In such an embodiment, the display device (124) may be a light emitting diode (OLED). In another embodiment, the device (100) includes a communication module (122) which is configured to communicate output data from a plurality of units of the housing unit (102) and a plurality of modules of the processor unit (112) to an external device (134). In such an embodiment, the external device (134) may be the display device (124). In another embodiment, the external device (134) may be a storage unit (128). In one embodiment, the communication module (122) may include one or more terrestrial and/or satellite networks interconnected to communicatively connect the device (100). In one example, the communication module (122) may be a private or public local area network (LAN) or Wide Area Network (WAN), such as the Internet. In another embodiment, the communication module (122) may include both wired and wireless communications according to one or more standards and/or via one or more transport mediums. In one example, the communication module (122) may include wireless communications according to one of the 802.11 or Bluetooth specification sets, or another standard or proprietary wireless communication protocol. In yet another embodiment, the communication module (122) may also include communications over a terrestrial cellular network, including, a GSM (global system for mobile communications), CDMA (code division multiple access), and/or EDGE (enhanced data for global evolution) network.
[0034] However, in one embodiment of the present disclosure, the device (100) may include the storage unit (128) which is coupled to the processor unit (112). The storage unit (128) is configured to store a set of predefined prediction rules (artificial learning/machine learning based rules) and a set of pre-defined health parameters. In another embodiment, the data is dynamically stored on the cloud.
[0035] Further, in one embodiment of the present disclosure the device (100) may be charged using wireless charger. In another embodiment, the device (100) may be connected to the external devices (134) via Bluetooth, Wi-Fi and the like.
[0036] FIG. 3 is a block diagram representing an exemplary embodiment of the device (100) for monitoring and prognosis of health of FIG. 1 in accordance with an embodiment of the present disclosure. Considering a non-limiting example, where in the operative position, the device (100) is places over chest part of the user X. The diaphragm (138) on the microphone (140) of the interface unit (104) obtains the plurality of auscultation signals form the user X. The plurality of auscultation signals are in the form of sound signals which are then received by the filtration unit (106). The filtration unit (106) includes the first level filter (106a) which filters the noise from the received plurality of auscultation signals. The filtration unit (106) also includes the amplifier (120) which increases the amplitude of the plurality of auscultation signals, and the second level filter (106b) reduces noise from the amplified signals. Hence amplified and filtered a plurality of analogue auscultation signals are obtained as an output of the filtration unit (106).
[0037] On the other hand, the strain sensor (108) obtains the plurality of analogue auscultation signals which are in the form of force applied by the skin variations. The strain sensor (108) converts the plurality of auscultation signals which are in the form of force into a plurality of electronic auscultation signal. The plurality of auscultation signals obtained as an output of filtration unit (106) and the plurality of auscultation signals received from the strain sensor (108) are given as an input to the converter unit (110). The converter unit (110) converts the plurality of analogue auscultation signals into the digital signals. The converted plurality of auscultation signals then integrated by integration module (114) to form a health dataset. The integrated signals then compared by the comparator module (116) by using the pre-defined health parameter and pre-defined health dataset stored in the storage unit (128) to detect health problems of the user. The prediction module (118) receives the output of the comparator module (116) and processes the signals by using artificial intelligence process to predict health issues in the body of the user X. In continuation with the aforementioned example, upon comparison, it is found that the health parameters of the user X are associated with the parameters which indicates probable heart attack. The prediction module, based on comparison result, will display the user X that the current health parameters may result into heart attack soon and user should take the further precaution or doctor consultation. The results are then communicated to the external devices (134) such as mobile, tablet and the like via communication module (122). Advantageously, if there is serious heart or lung problem occurred, the user may seek immediate medical help and future serious health conditions like heart attack or asthma attack may be avoided.
[0038] FIG. 4 is a block diagram of a computer or a server (300) for the device (100) for monitoring and prognosis of health in accordance with an embodiment of the present disclosure. The server includes processor(s) (302), and memory (306) operatively coupled to the bus (304).
[0039] The processor(s) (302), as used herein, means any type of computational circuit, such as, but not limited to, a microprocessor, a microcontroller, a complex instruction set computing microprocessor, a reduced instruction set computing microprocessor, a very long instruction word microprocessor, an explicitly parallel instruction computing microprocessor, a digital signal processor, or any other type of processing circuit, or a combination thereof.
[0040] The memory (306) includes a plurality of subsystems and a plurality of modules stored in the form of executable program which instructs the processor 302 to perform the method steps illustrated in FIG. 1. The memory (306) is substantially similar to the device (100) of FIG.1. The memory (306) has following submodules: an integration module (114), a comparator module (116) and a prediction module (118).
[0041] The integration module (114) is configured to receive the plurality of auscultation signals obtained from interface unit (104) of the device (100) and the one or more physiological signals obtained from the strain sensor (108) of the device (100). The integration module (114) is also configured to integrate such a plurality of digitized auscultation signals and physiological signals by the converter unit (110) to obtain an integrated health signal. The comparator module (116) is configured to compare the integrated health signals with historical health parameter datasets using a pre-trained model to obtain a comparative health dataset. The pre-trained model is trained with a plurality of health-issue-related datasets using a convolution neural network (CNN) technique in an iterative manner. The training of pre-trained model is done using a low power and low latency CNN based modelling. The trained CNN model is saved, and this model may be compared with input signal (sound signals and signals obtained from strain sensor (108) which is again modelled with the modified CNN. The prediction module (118) is configured to predict one or more health issues corresponding to the user based on the comparative health dataset, thereby performing monitoring and prognosis of health of the user. The one or more health issues is corresponding to at least one of heart dysfunction, asthma, pneumonia, and stress, of the user. The health signals from comparative dataset are processed using low power support vector machine (SVM) and artificial intelligence to predict heart and lung disorder.
[0042] Computer memory elements may include any suitable memory device(s) for storing data and executable program, such as read only memory, random access memory, erasable programmable read only memory, electrically erasable programmable read only memory, hard drive, removable media drive for handling memory cards and the like. Embodiments of the present subject matter may be implemented in conjunction with program modules, including functions, procedures, data structures, and application programs, for performing tasks, or defining abstract data types or low-level hardware contexts. Executable program stored on any of the above-mentioned storage media may be executable by the processor(s) (302).
[0043] FIG. 5 is a flow chart representing steps involved in a method (200) of operation of an obstetrician assistance device (100) in accordance with an embodiment of the present disclosure. The method (200) obtaining, by an interface unit (104), a plurality of auscultation signals of health parameters of a user in step (202). In one embodiment, a diaphragm (138) is a thin membrane that moves in reaction to external sound pressure variation. The diaphragm (138) is a key transducer component in converting the plurality of auscultation signals into electrical signals. The thinness of the diaphragm (138) makes them sensitive to vibrating air molecules in their immediate surroundings. In such another embodiment, when the device (100) is placed on the chest of the user, the bombardment of vibrating air molecule on a microphone (140) and diaphragm (138), causes the diaphragm (138) to move. This mechanical movement, in and out, from the resting position of the device (100) is converted into an electrical AC audio signal.
[0044] The method (200) also includes filtering, by a filtration unit, to eliminate noise signals from the plurality of auscultation signals obtained by the interface unit (204). In one embodiment, minimising, by a first level filter, the noise from the plurality of signals received from the interface unit. In such other embodiment, amplifying, by an amplifier the plurality of auscultation signals. In such other embodiments, reducing, by a second level filter, the noise from signals amplified by the amplifier. In one embodiment, using, first level filter as a low power modified CMOS (complementary metal-oxide semiconductor) programmable band pass filter and using, the second level filter as a programmable Butterworth filter. In the filtration unit, the first level filter minimises noise from the plurality of auscultation signals received from the microphone. The amplifier receives signals from the first level filter and amplifies the amplitude of the plurality of auscultation signals. The amplified plurality of auscultation signals are received by the second level filter which uses a Butterworth filter. The Butterworth filter filters the amplified signals to obtain pure form of the plurality of auscultation signals.
[0045] Furthermore, the method (200) includes converting, by a converter unit, to digitalise the plurality of auscultation signals filtered by filtration unit and the one or more physiological signals by strain sensor in step (208). In an exemplary embodiment, a 4-bit ADC may have a resolution whereas an 8-bit ADC may have low resolution. Thus, an analogue to digital converter takes an unknown continuous analogue signal and converts the signals into an “n”- bit binary number of 2n bits. In one embodiment, the converter may be a 24 -bit converter to digitalise the plurality of auscultation signals.
[0046] Furthermore, the method (200) includes integrating, by an integration module, to integrate the plurality of auscultation signals and the one or more physiological signals digitized by the converter unit to obtain an integrated health signal in step (210).
[0047] Furthermore, the method (200) includes comparing, by a comparator module, to compare the integrated health signals with historical health parameter datasets using a pre-trained model to obtain a comparative health dataset, where the pre-trained model is trained with a plurality of health-issue-related datasets using a convolution neural network (CNN) technique in an iterative manner in step (212).The trained CNN model is saved, and this model may be compared with input signal (sound signals and signals obtained from strain sensor which is again modelled with the modified CNN. In one embodiment, the CNN is a Deep Learning process which may take in an input signal, assign learnable weights and biases to various signals and be able to differentiate one from the other. Thus, the CNN model trains the pretrained model using low power and low latency.
[0048] Furthermore, the method (200) includes predicting, by a prediction module, to predict one or more health issues corresponding to the user based on the comparative health dataset, thereby performing monitoring and prognosis of health of the user, where the one or more health issues is corresponding to at least one of heart dysfunction, asthma, pneumonia, and stress of the user in step (214). The health signals from comparative dataset are processed using low power support vector machine (SVM) and artificial intelligence to predict heart and lung disorder.
[0049] The SVM are supervised learning models with associated learning algorithms that analyse data for classification and regression analysis. The regression analysis is a statistical process for estimating relationship between the data received and the historical data. SVM’s are one of the most robust prediction methods. The “prediction” refers to the output of a process after it has been trained on a historical dataset and applied to new data when forecasting a particular outcome. In an exemplary embodiment, artificial intelligence (AI) platform prediction manages computing resources in the cloud to run the trained models. The predictions may be requested from the trained models and may get predicted target values for them. In one embodiment, the process to make predictions in the cloud includes exporting of trained model as artifacts that may be deployed to AI platform prediction. The processes also include creating a model resource in AI platform prediction and then create a model version from the saved model. For online prediction, the service runs the saved model and returns the requested predictions as the response message for the call.
[0050] Further, in one embodiment, the method (200) includes communicating, by a communication module to communicate output data from a plurality of units of the housing unit and a plurality of modules of the processor unit to an external device. In another embodiment, the method (200) includes a display device configured to display prediction of the one or more health issues corresponding to the user, where the display device includes an organic light-emitting diode (OLED). In another embodiment, the method (200) includes charging, by a charging unit configured to charge a battery to power up a plurality of units of the housing unit. In such an embodiment, the method (200) includes, power boosting, by a booster circuit for boosting power of the processor unit by using the charging unit. In another embodiment, the method (200) includes storing, by a storage unit configured to store a set of predefined prediction rules and a set of pre-defined health parameters.
[0051] Further, from a technical effect point of view, the implementation time required to perform the method steps included in the present disclosure is very minimal, thereby the device is easy to operate and functions efficiently.
[0052] Various embodiments of the present disclosure enable health monitoring and prediction of health problems in relation to heart, lung and the like. The device in the present disclosure, may be operated without any help of medical expert. The device enables a user, to operate without going to hospital physically. Thus, heart and cardiac health may be monitored at home. The device predicts the future problems in relation to heart and lung so that the user may seek medical help in early stage and avoid serious health situations.
[0053] While specific language has been used to describe the disclosure, any limitations arising on account of the same are not intended. As would be apparent to a person skilled in the art, various working modifications may be made to the method in order to implement the inventive concept as taught herein.
[0054] The figures and the foregoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, order of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts need to be necessarily performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples.
,CLAIMS:1. A device for monitoring and prognosis of health (100), wherein the device (100) comprises:
a housing unit (102) comprising:
an interface unit (104) configured to obtain a plurality of auscultation signals of a user;
a filtration unit (106) operatively coupled to the interface unit (104), wherein the filtration unit (106) configured to eliminate noise signals from the plurality of auscultation signals;
a strain sensor (108) configured to sense one or more physiological signals corresponding to stress level of the user by measuring skin resistance variations; and
a converter unit (110) operatively coupled with the filtration unit (106) and the strain sensor (108), wherein the converter unit (110) configured to digitalize the plurality of auscultation signals filtered by the filtration unit (106) and the one or more physiological signals sensed by the strain sensor (108); and
a processor unit (112) communicably coupled to the housing unit (102), wherein the processor unit (112) comprises:
an integration module (114) configured to integrate the plurality of auscultation signals and the one or more physiological signals digitized by the converter unit (110) to obtain an integrated health signal;
a comparator module (116) configured to compare the integrated health signals with historical health parameter datasets using a pre-trained model to obtain a comparative health dataset,
wherein the pre-trained model is trained with a plurality of health-issue-related datasets using a convolution neural network (CNN) technique in an iterative manner; and
a prediction module (118) configured to predict one or more health issues corresponding to the user based on the comparative health dataset, thereby performing monitoring and prognosis of health of the user, wherein the one or more health issues is corresponding to at least one of heart dysfunction, asthma, pneumonia, and stress, of the user.
2. The device (100) as claimed in claim 1, wherein the filtration unit (106) comprises an amplifier (120) to amplify the plurality of auscultation signals.
3. The device (100) as claimed in claim 1, wherein the filtration unit (106) comprises:
a first level filter (106 a) configured to minimize the noise from the plurality of signals received from the interface unit (104); and
a second level filter (106 b) configured to reduce the noise from signals amplified by the amplifier (120).
4. The device (100) as claimed in claim 3, wherein the first level (106 a) filter is a low power modified CMOS (complementary metal-oxide semiconductor) programmable band pass filter and the second level filter (106 b) is a programmable Butterworth filter.
5. The device (100) as claimed in claim 1, comprises a communication module (122) to communicate output data from a plurality of units of the housing unit (102) and a plurality of modules of the processor unit (112) to an external device (134).
6. The device (100) as claimed in claim 1, comprises a display device (124) configured to display prediction of the one or more health issues corresponding to the user, wherein the display device (124) comprises an organic light-emitting diode (OLED).
7. The device (100) as claimed in claim 1, comprises a charging unit (130) configured to charge a battery (136).
8. The device (100) as claimed in claim 7, comprises a booster circuit (132) powered by the battery (136), wherein the booster circuit (132) boosts power received from the charging unit (130) and provide boosted power to the processor unit (112).
9. The device (100) as claimed in claim 1 comprises a storage unit (128) configured to store a set of predefined prediction rules and a set of pre-defined health parameters.
10. A method (200) for monitoring and prognosis of health comprises steps of:
obtaining, by an interface unit, a plurality of auscultation signals of health parameters of a user; (202)
filtering, by a filtration unit, to eliminate noise signals from the plurality of auscultation signals obtained by the interface unit; (204)
sensing, by a strain sensor, to sense one or more physiological signals corresponding to stress level of the user by measuring skin resistance variations; (206)
converting, by a converter unit, to digitalise the plurality of auscultation signals filtered by filtration unit and the one or more physiological signals by strain sensor; (208)
integrating, by an integration module, to integrate the plurality of auscultation signals and the one or more physiological signals digitized by the converter unit to obtain an integrated health signal; (210)
comparing, by a comparator module, to compare the integrated health signals with historical health parameter datasets using a pre-trained model to obtain a comparative health dataset, wherein the pre-trained model is trained with a plurality of health-issue-related datasets using a convolution neural network (CNN) technique in an iterative manner; (212) and
a predicting, by a prediction module, to predict one or more health issues corresponding to the user based on the comparative health dataset, thereby performing monitoring and prognosis of health of the user, wherein the one or more health issues is corresponding to at least one of heart dysfunction, asthma, pneumonia, and stress of the user. (214)

Dated this 28th day of April 2022

Signature

Jinsu Abraham
Patent Agent (IN/PA-3267)
Agent for the Applicant