Claims:1. A method for non-invasive detection of diabetic disease in an individual, the method comprising:
receiving, by a processor, a probe Photo-Plethysmogrpahy signal (PPG) from an individual under test, wherein the individual under test is to be classified as at least one of a normal or pre-diabetic or diabetic individual;
preprocessing by the processor the probe PPG signal, wherein preprocessing includes,
amplifying by the processor the probe PPG signal to produce a probe amplified PPG signal,
removing by the processor noise from the probe amplified PPG signal to produce a probe filtered PPG signal,
sampling by the processor the probe filtered PPG signal into a predefined intervals to produce a probe preprocessed PPG signal;
analyzing by the processor, the probe pre-processed PPG signal, using harmonic analysis for extracting a first set of feature parameters from the probe pre-processed PPG signal; and
matching, by the processor, the first set of feature parameters with a second set of feature parameters extracted corresponding to a control group of individuals, wherein the control group of individuals include individuals with previously known classification as normal, pre-diabetic or diabetic, by using machine learning techniques to classify the individual under test as one of the normal, or pre-diabetic, or diabetic individual.

2. The method of claim 1 further comprises a step for extraction of the second set of feature parameter corresponding to the control group of individual comprises:
receiving by the processor a training PPG signal from each individual of the control group of individuals;
pre-processing by the processor, the training PPG signal, wherein the pre-processing includes,
amplifying, by the processor, the training PPG signal to produce an training amplified PPG signal;
removing, by the processor, noise from the training amplified PPG signal to produce a filtered PPG signal; and
sampling by the processor, the training filtered PPG signal into predefined intervals to produce a training preprocessed PPG signal; and
analyzing, by the processor, the training pre-processed PPG signal, using harmonic analysis for extracting the training set of feature parameters from the training preprocessed PPG signal.

3. The method of claim 1 further comprising, receiving the PPG signal using a sensor for monitoring arterial pulse palpation of the individual under test, adapting an arterial pulse data by the monitoring and transmitting the arterial pulse data to at least one central server, remote device or cloud server, wherein the arterial pulse data comprises the probe PPG signal.

4. The method of claim 2 further comprising, receiving the PPG signal, using a sensor for monitoring arterial pulse palpation of each individual from the control group of individuals, adapting an arterial pulse data from the monitoring and transmitting the arterial pulse data to at least one remote device or a cloud server, wherein the arterial pulse data comprises the training PPG signal.

5. The method of claim 1, wherein the first set of feature parameters comprises nine level of harmonics for the first PPG signal, and the second set of feature parameters comprises nine level of harmonics of the second PPG signal wherein harmonic analysis includes the determining amplitude and phase angle for each harmonic of the waveform using Fourier Series Coefficients.

6. The method of claim 1 further comprising:
monitoring, by the processor a pulse wave velocity of the individual under test and;
matching by the processor the pulse wave velocity of the individual under test at different time instant to estimate a health severity index for the individual under test.

7. The method of claim 1 further comprising:
monitoring by the processor a pulse wave transition time for the individual under test; and
matching by the processor, the pulse wave transition time for the individual under test at different time instant to estimate the progression risk of diabetes disease for the individual under test.

8. The method of claim 1 further comprising:
receiving one or more physiological parameters for the individual under test, wherein the one or more physiological parameters comprises a blood glucose level, an electrocardiogram, a pulse wave velocity or a pulse transmission time; and
estimating by a processor, a vascular system risk for the individual under test using a decision fusion approach, wherein decision fusion approach includes fusing an output of matching of the second set of feature parameters with the first set of feature parameters with the one or more physiological parameters.

9. A device (400) for non-invasive detection of diabetic individual, the device (400) comprising:
a sensor to acquire a probe PPG signal by monitor an arterial pulse palpation of an individual;
a processor ; and
a memory coupled to the processor, wherein the processor executes computer readable instructions stored in the memory to:
receive, by the processor, a probe PPG signal from an individual under test, wherein the individual under test is to be classified as at least one of a normal or pre-diabetic or diabetic individual;
pre-process, by the processor, the probe PPG signal, wherein the pre-processing includes,
amplifying by the processor the probe PPG signal to produce a probe amplified PPG signal,
removing by the processor noise from the probe amplified PPG signal to produce a probe filtered PPG signal,
sampling by the processor the probe filtered PPG signal into a predefined intervals to produce a probe preprocessed PPG signal;
analyze by the processor, the probe pre-processed PPG signal, using harmonic analysis for extracting a first set of feature parameters from the probe pre-processed PPG signal; and
match, by the processor, the first set of feature parameters with a second set of feature parameters extracted corresponding to a control group of individuals, wherein the control group of individuals include individuals with previously known classification as normal, pre-diabetic or diabetic, by using machine learning techniques to classify the individual under test as one of the normal, or pre-diabetic, or diabetic individual.

10. The device of claim 9 further comprising computer program code comprising instructions executable by the processor to:
receive by a processor a training PPG signal from each individual of the control group of individuals;
preprocess by the processor, the training PPG signal, wherein the preprocessing includes,
amplifying, by the processor, the training PPG signal to produce an training amplified PPG signal;
removing, by the processor, noise from the training amplified PPG signal to produce a filtered PPG signal; and
sampling by the processor, the training filtered PPG signal into predefined intervals to produce a training preprocessed PPG signal; and
analyze, by the processor, the training pre-processed PPG signal, using harmonic analysis for extracting the training set of feature parameters from the training preprocessed PPG signal.

11. The device of claim 9 wherein the sensor monitors the arterial pulse palpation from a radial artery.

12. The device of claim 9 further comprising computer program code comprising instructions executable by the processor to:
continuously monitor the value of a pulse wave velocity (PWV) for the individual under test and;
comparing the pulse wave velocity at two distinct time instants to determine the health severity index and risk of progression of diabetes disease.

13. The device of claim 9 further comprising computer program code comprising instructions executable by the processor to:
continuously monitor the value of a pulse wave transition time (PTT) for the individual under test and;
comparing the pulse wave transition time at two distinct time instants to determine the health severity index and risk of progression of diabetes disease.

14. The device of claim 9 wherein the sensor monitoring the arterial pulse palpation of the individual under test and transmitting an arterial pulse data to a central server, at least one remote device or a cloud server, wherein the arterial pulse data comprises the probe PPG signal.

15. The device of claim 9 wherein the sensor monitoring the arterial pulse palpation of the individual under test and transmitting an arterial pulse data to at least one central server remote device or cloud server, wherein the arterial pulse data comprises information regarding the training PPG signal.

16. The device of claim 9 further operatively coupled with a mobile device wherein the processor may be configured to display an information corresponding to the classification of the individual under test as a normal, a pre-diabetic or a diabetic individual.

17. The device of claim 12 further operatively coupled with a mobile device wherein the processor is configured to display an information corresponding to the health severity index and progression risk of the individual under test.

18. The device of claim 13 further operatively coupled with a mobile device wherein the processor is configured to display an information corresponding to the health severity index and progression risk of the individual under test.

19. The device of claim 9 further comprising computer program code comprising instructions executable by the processor and configured to:
receive one or more physiological parameters for the individual under test, wherein the one or more physiological parameters comprises a blood glucose level, an electrocardiogram, a pulse wave velocity or a pulse transmission time; and
estimate, a vascular condition risk for the individual under test using a decision fusion approach, wherein decision fusion approach includes fusing an output of matching of the second set of feature parameters with the first set of feature parameters with the one or more physiological parameters.
, 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:
METHOD AND DEVICE FOR CONTINUOUS NON-INVASIVE DETECTION OF DIABETES DISEASE

Applicant
Tata Consultancy Services Limited
A company Incorporated in India under the Companies Act, 1956
Having address:
Nirmal Building, 9th floor,
Nariman point, Mumbai 400021,
Maharashtra, India

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

[001] The present application generally relates to screening tool for diabetes. More particularly, but not specifically, the invention provides a method and device for adaptive and autonomous obtaining arterial volumetric pulsations of blood which may be analyzed for screening of diabetes.

BACKGROUND OF THE INVENTION

[002] Diabetes consist of malfunction of glucose-insulin regulatory system leading to the onset various complications. As per a prediction made World Health Organizations report (WHO/NMH/MNC/03.1), there will be at least 350 million people in the world with type 2 diabetes by the year 2030. Therefore it becomes essential to screen people for diabetes on a regular basis. Further it is uncertain that whether such screen may be done at a population wide basis or just for people who can be shown to have high risk. It is also uncertain at what age the diabetes screening program should be initiated. Furthermore such screening also proves beneficial for assessment of long term health condition risks like type 2 diabetes, heart disease, hypertension, stroke, kidney disease, some forms of dementia such as Alzheimer’s and so on. Also, continuous monitoring of diabetes patients can aid in assisting the short and long term complication risks as well.

[003] A majority of existing solutions for such monitoring relies on techniques such as C-peptide test, fasting plasma glucose test, GAD antibodies test, Hba 1c test, oral glucose tolerance test, type-2 diabetes indication test. It should be noted that most of the above mentioned technique are either invasive or minimal invasive (figure prick) in nature. Further, based on the blood glucose level an individual will be mapped with normal, pre-diabetic or diabetic. Furthermore the sensors used in the above techniques may be uncomfortable for the patient and are typically used no more than three four times a day.

[004] The prior art literature does not explore non-invasive methods of diabetic screening further none of the prior art discloses methods to continuously monitor the vulnerability of an individual towards diabetes so as to determine the severity index of the disease and estimate the health condition risk due to diabetes.
[005] Some prior art have vaguely considered Pulse wave analysis to be a simple, noninvasive and informative technique for arterial assessment in which the central arterial blood pressure can be estimated from the Brachial Blood Pressure. However, the prior literature does not reach a consensus regarding the validity of the technique. Instead, Murgo et al (1980;62:105–116.) teaches that the shape of the arterial pulse will be affected by the changes in the peripheral circulation or alterations in cardiac function.

[006] The above-mentioned prior art lacks the advantages of a continuous monitoring system that is capable of being worn for continual periods. Therefore, there is a need to provide a method and device for the continuous, non-invasive monitoring and estimation of the blood pressure.

[007] Prior art literature have illustrated various method of determining and screening of diabetes, however, continuous and non-invasive screening of diabetes is still considered as one of the biggest challenges of the technical domain.

SUMMARY OF THE INVENTION

[008] Before the present methods, systems, and hardware enablement are described, it is to be understood that this invention is not limited to the particular systems, and methodologies described, as there can be multiple possible embodiments of the present invention which are not expressly illustrated in the present disclosure. It is also to be understood that the terminology used in the description is for the purpose of describing the particular versions or embodiments only, and is not intended to limit the scope of the present invention which will be limited only by the appended claims.

[009] The present application provides a method and system for non-invasive detection of diabetic individual is disclosed. According to an embodiment the invention discloses a method for non-invasive detection of a diabetic individual. A probe Photo-Plethysmogrpahy signal (hereinafter referred to as PPG) is received from an individual under test, wherein the individual under test is to be classified as at least one of a normal or pre-diabetic or diabetic individual. The probe PPG signal is preprocessed. Preprocessing includes, amplifying the probe PPG signal to produce a probe amplified PPG signal, removing noise from the probe amplified PPG signal to produce a probe filtered PPG signal and sampling the probe filtered PPG signal into a predefined intervals to produce a probe preprocessed PPG signal. The probe pre-processed PPG signal is analyzed using harmonic analysis for extracting a first set of feature parameters from the probe pre-processed PPG signal. The first set of feature parameters with a second set of feature parameters extracted corresponding to a control group of individuals, wherein the control group of individuals include individuals with previously known classification as normal, pre-diabetic or diabetic, by using machine learning techniques to classify the individual under test as one of the normal, or pre-diabetic, or diabetic individual.

[0010] The present application, in an embodiment provides a device (400) for non-invasive detection of diabetic individual; A device (400) for non-invasive detection of diabetic individual; said device (400) comprising: a strap to be worn by an individual; a sensor to acquire a PPG signal by monitor the arterial pulse palpation or pulse signal of an individual; a processor ; and a memory coupled to the processor, wherein the processor executes computer readable instructions stored in the memory to receive, a probe PPG signal from an individual under test. receiving, by the processor, a probe PPG signal from an individual under test, wherein the individual under test is to be classified as at least one of a normal or pre-diabetic or diabetic individual; preprocessing by the processor the probe PPG signal, wherein preprocessing includes, amplifying by the processor the probe PPG signal to produce a probe amplified PPG signal, removing by the processor noise from the probe amplified PPG signal to produce a probe filtered PPG signal, sampling by the processor the probe filtered PPG signal into a predefined intervals to produce a probe preprocessed PPG signal; analyzing by the processor, the probe pre-processed PPG signal, using harmonic analysis for extracting a probe set of feature parameters from the probe pre-processed PPG signal; and matching, by the processor, the probe set of feature parameters with a training set of feature parameters extracted corresponding to a control group of individuals, wherein the control group of individuals include individuals with previously known classification as normal, pre-diabetic or diabetic, by using machine learning techniques to classify the individual under test as one of the normal, or pre-diabetic, or diabetic individual.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The foregoing summary, as well as the following detailed description of preferred embodiments, are better understood when read in conjunction with the appended drawings. For the purpose of illustrating the invention, there is shown in the drawings exemplary constructions of the invention; however, the invention is not limited to the specific methods and system disclosed. In the drawings:

[0012] Figure 1: shows a flow chart illustrating a method for implementation of a training phase of the disclosed invention.

[0013] Figure 2: shows a flow diagram illustrating the method for implementing a probe phase of the disclosed invention.

[0014] Figure 3: shows a flow diagram illustrating the matching of extracted features and classification of individual as normal, pre-diabetic or diabetic.

[0015] Figure 4: shows a schematic diagram illustrating various parts of a device for monitoring and estimating diabetic condition of an individual, in accordance with an embodiment of the invention.

[0016] Figure 5: shows an exemplary implementation of the device for monitoring and estimating diabetic condition of an individual, in accordance with one embodiment of the invention.

[0017] Figure 6: shows an exemplary waveform representation of the received PPG signal for the Control Group and Individual under test, in accordance with an embodiment of the present disclosure.

[0018] Figure 7: shows an exemplary representation of matching of the feature parameters extracted from the control group and individual under test.

DETAILED DESCRIPTION OF THE INVENTION

[0019] Some embodiments of this invention, illustrating all its features, will now be discussed in detail.

[0020] The words "comprising," "having," "containing," and "including," and other forms thereof, are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items.

[0021] It must also be noted that as used herein and in the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. Although any systems and methods similar or equivalent to those described herein can be used in the practice or testing of embodiments of the present invention, the preferred, systems and methods are now described.

[0022] The disclosed embodiments are merely exemplary of the invention, which may be embodied in various forms.

[0023] The elements illustrated in the Figures inter-operate as explained inmore detail below. Before setting forth the detailed explanation, however, it is noted that all of the discussion below, regardless of the particular implementation being described, is exemplary in nature, rather than limiting. For example, although selected aspects, features, or components of the implementations are depicted as being stored in memories, all or part of the systems and methods consistent with the attrition warning system and method may be stored on, distributed across, or read from other machine-readable media.

[0024] The techniques described above may be implemented in one or more computer programs executing on (or executable by) a programmable computer including any combination of any number of the following: a processor, a storage medium readable and/or writable by the processor (including, for example, volatile and non-volatile memory and/or storage elements), plurality of input units, and plurality of output devices. Program code may be applied to input entered using any of the plurality of input units to perform the functions described and to generate an output displayed upon any of the plurality of output devices.

[0025] Each computer program within the scope of the claims below may be implemented in any programming language, such as assembly language, machine language, a high-level procedural programming language, or an object-oriented programming language. The programming language may, for example, be a compiled or interpreted programming language. Each such computer program may be implemented in a computer program product tangibly embodied in a machine-readable storage device for execution by a computer processor.

[0026] Method steps of the invention may be performed by one or more computer processors executing a program tangibly embodied on a computer-readable medium to perform functions of the invention by operating on input and generating output. Suitable processors include, by way of example, both general and special purpose microprocessors. Generally, the processor receives (reads) instructions and data from a memory (such as a read-only memory and/or a random access memory) and writes (stores) instructions and data to the memory. Storage devices suitable for tangibly embodying computer program instructions and data include, for example, all forms of non-volatile memory, such as semiconductor memory devices, including EPROM, EEPROM, and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; and CD-ROMs. Any of the foregoing may be supplemented by, or incorporated in, specially-designed ASICs (application-specific integrated circuits) or FPGAs (Field-Programmable Gate Arrays). A computer can generally also receive (read) programs and data from, and write (store) programs and data to, a non-transitory computer-readable storage medium such as an internal disk (not shown) or a removable disk.

[0027] In one embodiment, the device may be implemented using a network, the network may be a wireless network, a wired network or a combination thereof. The network can be implemented as one of the different types of networks, such as intranet, local area network (LAN), wide area network (WAN), the internet, and the like. The network may either be a dedicated network or a shared network. The shared network represents an association of the different types of networks that use a variety of protocols, for example, Hypertext Transfer Protocol (HTTP), Transmission Control Protocol/Internet Protocol (TCP/IP), Wireless Application Protocol (WAP), and the like, to communicate with one another. Further the network may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, and the like.

[0028] Any data disclosed herein may be implemented, for example, in one or more data structures tangibly stored on a non-transitory computer-readable medium. Embodiments of the invention may store such data in such data structure(s) and read such data from such data structure(s).

[0029] In one aspect, a system and method for continuously or periodically monitoring and estimating diabetic condition of a person is disclosed. The invention disclosed herein may be used for detecting a diabetic individual thereby making a device used for screening diabetes disease. The system continuously or periodically monitors the arterial pulse palpation in a non-invasive manner to detect diabetes disease.

[0030] In another aspect the invention disclosed herein may be used for continuously monitoring physiological parameters and estimate a health severity index of a diabetic individual. The invention discloses acquiring or collecting a pulse signal by using an existing device or a specially designed device.

[0031] The present application provides a method for continuously or periodically monitoring and estimating diabetic condition of a person. The invention disclosed herein may be used for detecting a diabetic individual thereby making it a device for screening diabetes disease. The application discloses continuous or periodic monitoring an arterial pulse palpation in a non-invasive manner to detect diabetes disease.

[0032] In yet another aspect the invention disclosed herein may acquire other information like his medication or fitness activity in addition to the physiological signal like the arterial pulse signal and estimate a health severity index of a diabetic individual. The invention discloses acquiring or collecting the fitness activity or medication signal by embedded device or otherwise. The health severity index may be determined based on a pulse wave velocity (PWV) or the pulse wave transmission time (PTT) for the individual under test.

[0033] The pulse wave velocity or the pulse wave transmission time may be calculated by using a more than one electrode system on the same peripheral organ which is used for calculating the probe set of feature parameters or the training set of feature parameters and stored as a third set of feature parameters. This third set of feature parameters is used for determining the health severity of the individual under test.

[0034] Further the invention may be used for determining the progression risk of diabetes disease using the feature parameters and the pulse transition time or pulse wave velocity

[0035] In an embodiment the progression risk of diabetes disease is estimated using a pulse transition time (PTT) or a pulse wave velocity (PWV) of the individual under test

[0036] The application in another aspect discloses a method and device for non-invasive detection of diabetes disease and screening of diabetic individuals. The invention provides a non-invasive method for classifying an individual as normal, pre diabetic or diabetic.

[0037] In yet another aspect the application discloses a method and device for continuous or periodic non-invasive monitoring of arterial pulse palpation to detect diabetes disease. The method and device disclosed herein may be used for detection of individuals with diabetes disease.

[0038] In one embodiment, the non-invasive monitoring technique may include using a Plethysmogrpahy device. Various plethysmogrpahy techniques are in existence and each of the plethysmogrpahy technique is used to measure a change in blood volume. In one aspect the present invention employs Photo-Plethysmogrpahy (hereinafter referred to as PPG) devices for non-invasive monitoring of arterial pulse palpation.

[0039] In an embodiment the PPG signal may be acquired by a sensor attached to a peripheral artery of an individual. The PPG in general, device contains a light source and a detector to detect cardiovascular pulse wave that propagates through the body of a subject. Meaning thereby that the PPG signal represents the blood movement in the vessel that flows from the heart to fingertips (via radial artery) and toes through blood vessels.

[0040] In an embodiment the disclosed method for detection of diabetic individual and diabetes disease may be implemented in two phases, 1) Training Phase and 2) Probe Phase. The working of the invention in the two phases in an exemplary embodiment are explained in greater detail in the following paragraphs.

[0041] Referring to Figure 1 is a flow chart illustrating the training phase for training the invention disclosed herein. The training phase starts at step 102, where a training PPG signal is acquired by the system. In an embodiment the training PPG signal may be acquired by monitoring the arterial pulse palpation of a control group of individuals. In an embodiment the control group of individuals includes individuals with known classification as normal, pre diabetic or diabetic.

[0042] At step 104 the training PPG signal is processed to produce a training preprocessed PPG signal. In an embodiment, preprocessing of the training PPG signal includes amplifying the training PPG signal, to produce a training amplified PPG signal, filtering any noise from the training amplified PPG signal to produce a training filtered PPG signal and sampling the training filtered PPG signal at a predefined frequency to produce a training preprocessed PPG signal.

[0043] In an embodiment the training PPG signal is amplified due to the extremely low magnitude of the initially acquired training PPG signal. In another embodiment the filtering of the training amplified PPG signal is performed to remove noise from the training amplified PPG signal. In another embodiment the sampling of the training filtered PPG signal may be performed by using an Analog digital convertor (ADC). In yet another an embodiment the predefined frequency for sampling the training filtered PPG signal may be 60Hz.

[0044] At step 106 the training preprocessed PPG signal undergoes harmonic analysis. In an embodiment the training preprocessed PPG signal may be represented by the equation X(t) = {x1, x2, x3….xn}. The harmonics of the training preprocessed PPG signal may be analyzed such that such that coefficient {xn} is related to X(t) as per the following mathematical equation

[0045] Further, the training preprocessed PPG signal X(t) is analyzed by implementing the following finite (fourier) series as shown in Equation 1 to the harmonics of the training preprocessed PPG signal.

……Equation 1

[0046] The Fourier coefficients (An and Bn) of the above mentioned finite series may be calculated by using the following equations 2 and 3
……..Equation 2 and 3
wherein the symbol ? represents angular frequency and ts represents sampling time interval.

[0047] At the step 108 a first set of feature parameters are extracted from the training preprocessed PPG signal using harmonic analysis. A sinewave components of a periodic waveform of the training preprocessed PPG signal, having a period which is an integer fraction of a fundamental period of the periodic waveform. In other words, each sinewave component occurs at a frequency which is an integer multiple of a fundamental frequency of the periodic waveform. The sinewave components are called harmonics (H). In an embodiment an amplitude (Ampn) and a phase angle (Pn) of each harmonic of the pre-processed PPG signal maybe generated by using the mathematical formula written in equation no….. 4and 5. In an embodiment the amplitude (Ampn) and phase angle (Pn) of each harmonic of the pre-processed PPG signal maybe generated by using the mathematical formula as mentioned in equation no… Equation 3 and 4

……. Equation 3 and 4
The value of Ampn and Pn are used to calculate the feature parameters for the harmonic and the feature parameters may be termed as H1, H2… for each value of n.

[0048] In one exemplary embodiment, nine level of harmonic frequency component of the training preprocessed PPG signal (H1, H2... H9) may be extracted using harmonic analysis. The nine level of harmonic component are used as feature parameters.

[0049] In another embodiment the PPG spectra from the averages of all the pluses during the entire measurement period for the amplitude proportion of the nth harmonic according to Ampn/Amp0 X 100% for n=1 to 10 is calculated during harmonic analysis, where Ampn is defined as the amplitude of the nth harmonic of the PPG pulse and Amp0 is the DC component or mean value of the PPG pulse spectrum.

[0050] At step 110, the extracted first set of feature parameters are stored for future processing by the system.

[0051] Referring to Fig 2, a flow diagram for the probe phase illustrating the implementation of the invention for classifying individuals as normal, pre-diabetic or diabetic. The probe phase starts at step 202, wherein a probe PPG signal is acquired by monitoring the arterial pulse palpation of an individual under test, wherein the individual under test is an individual to be classified as normal, pre-diabetic or diabetic.

[0052] At step 204 the probe PPG signal is processed to produce a probe preprocessed PPG signal. In an embodiment, preprocessing of the probe PPG signal includes amplifying the probe PPG signal, to produce a probe amplified PPG signal, filtering any noise from the probe amplified PPG signal to produce a probe filtered PPG signal and sampling the probe filtered PPG signal at a predefined frequency to produce a probe preprocessed PPG signal.

[0053] In an embodiment the probe PPG signal is amplified due to the extremely low magnitude of the initially acquired training PPG signal. In another embodiment the filtering of the probe amplified PPG signal is performed to remove noise from the probe amplified PPG signal. In another embodiment the sampling of the probe filtered PPG signal may be performed by using an Analog digital convertor (ADC). In yet another an embodiment the predefined frequency for sampling the filtered PPG signal may be 60Hz.

[0054] At step 206 the probe preprocessed PPG signal undergoes harmonic analysis. In an embodiment the harmonic analysis of the probe preprocessed PPG signal is harmonically analyzed using similar methodology as the training pre-processed PPG signal and a second set of feature parameters are extracted at step 208.

[0055] At the step 210 the second set of Feature parameter are stored for further processing by the system.

[0056] Referring to Figure 3 is a flow diagram illustrating the steps matching the feature parameters extracted during the training phase and probe phase to classify the individual under test as normal, pre diabetic or diabetic.

[0057] Element 302 represents the first set of feature parameters and element 304 represents the second set of feature parameters. The first set of feature parameters and the second set of feature parameters are matched using at least one machine learning techniques at step 306 to classify an individual under test as Normal (308), pre-diabetic (310) or diabetic (312).

[0058] In an example the number of individuals forming the part of the control group is U, the number of time an individual comes for trail is T and length of harmonic corresponding to the time duration for which the PPG signal is acquired is referred to as N.

[0059] Once the feature parameters for matching are selected, the system is trained using at least one machine learning technique and the first set of parameter is matched with the second set of parameter to classify the individual under test as normal, pre-diabetic or diabetic. For example the machine learning techniques comprises artificial neural networks (ANN), logistic regression, support vector machines (SVM) and the like.

[0060] It will be appreciated by a person skilled in the art that after undergoing the training phase once the system is trained, the probe phase may be implemented independent of the training phase.

[0061] Further in an embodiment the method disclosed herein may further include calculating PWT and PTT signal on the same peripheral artery which is used for calculating the probe set of feature parameters or the training set of feature parameters. This value of PWT signal and PPT is stored as a third set of feature parameters. This third set of feature parameters may be used for determining a health severity index or progression risk of diabetes disease. The estimation of health severity index may include steps of analyzing the PWT or PTT signals for the individual under test at different time intervals to determine if the health of the individual is improved or further degraded or no deviation has occurred.

[0062] In an embodiment PTT/PWV of an individual during a first time instant PTT=x msec. During next measurement data collected indicates PPT =y msec Where y= x??x in that such a case it may be adjudged that
if xy == Health has to be improved
x=y no improvement

[0063] In another embodiment, the physiological data or signals may be captured by using non-invasive method using one or more sensors. In an embodiment the sensor may be attached to a wrist or a fingertip or any part of body of the human, where the atrial pulse can be sensed (peripheral organs). For example, a wrist watch or a wristband or a textile material such as cuff to measure the arterial pulse palpation signals or biological parameter, or a ring or finger-cap to measure arterial pulse at the finger-tip. In one aspect, the arterial pulse palpation signals are captured, from at least one external sensor, for a predetermined ultra-short or short duration.

[0064] The invention disclosed herein may further be used to display the results of the classification of the individual under test using a mobile device. The mobile device may be a general mobile device such as a smart phone electronically coupled with the system disclosed herein or may be a specialized device electronically coupled to the system disclosed herein and configured to display the classification of the individual under test.

[0065] In another embodiment the mobile device may further be used to display the processed information regarding the health severity index

[0066] The disclosed invention may further be incorporated such that the individual user test may store information or preferences as to triggering an alert to a predefined point of contact. In an example an alert may be triggered when the individual is adjudged above a predefined threshold on the severity index. In another example the alert may be triggered when the individual under test is determined as diabetic.

[0067] The alert may include sending a distress call to a predefined phone or sending a message to a predefined phone number.

[0068] According to an embodiment of the present invention, the device 400 includes a strap 402, a PPG sensor 404 present on the strap 402, a processor 406, a memory 408 a data-storage 410 and a display device 412 as shown in Fig. 4. The processor 406 is electronically coupled with the memory 408, the data storage 410 and the display device 412. The processor 406 is configured to take input from the PPG sensor 404 to generate a pre-processed PPG signal. The processor 406 is further configured to harmonically analyze the pre-processed PPG signal to extract a set of feature parameters. The extracted feature parameters may be stored in the data storage element 410 of the device and may be used for further processing. The memory element may store various programed instructions to be performed by the processor 406. The processor 406 is further configured to match the first set of feature parameters from the training phase with the second set of feature parameters from the probe phase to by implementing machine learning techniques to classify the individual under test as normal, pre-diabetic or diabetic.

[0069] The PPG sensor 404 may acquire the PPG signal by monitoring the arterial pulse palpation at peripheral organs. In an embodiment the pulse palpation may be monitored on the wrist. In another embodiment the arterial pulse may be monitored at the tip of a finger.

[0070] In an exemplary embodiment referring to figure 5 the device 400 maybe a wearable device where the sensor 404 on the strap 402 acquires the PPG signal by monitoring the pulse palpation on the wrist.

[0071] In an embodiment referring to Fig 6, the PPG signals acquired by the sensor 404 may be depicted in the form of an amplitude versus Sample in second graph for the control group of individuals during the training phase and individual under test during the probe phase. Since the PPG signal may be low in magnitude therefore the preprocessing module to the magnitude of 1mV the processor 406 is configured to amplify the acquired PPG signal.

[0072] Figure 7 illustrates an exemplary representation of comparison of the feature parameters extracted during the working of the disclosed method using SVM technique, such that
…. Equation 5
The graph in figure 7 illustrates the comparison between 9 harmonics (H1, H2 … H9) and mean and standard error mean of features parameters extracted during the harmonic analysis of the preprocessed PPG signal extracted from the control group and the individual under test.

[0073] In the above mentioned equation no…. 5, Y2= 1 for Normal Individuals, 0 for diabetic individuals, x1, x2....xn are the features with high impact coefficient, f(x1, x2….xn) is a representation of feature parameters.

[0074] In yet another embodiment the processor 406 may further be configured to transmit the acquired arterial pulse signal to central server or cloud or remote device where the arterial pulse signal may be stored for further processing.

[0075] In another embodiment the invention may use other physiological parameters collected using invasive or imaging technique like blood glucose level or echocardiogram, pulse wave transmission risk to determine the cardiac risk to the individual under test. The processor 406 may further be configured to fuse the diabetes information generated after matching the training feature parameters with the probe feature parameters, with other physiological parameters to estimate cardiac risk to the individual under test.

[0076] In another embodiment of the present invention, the processor 406 may be configured to match the first set of feature parameters with the second set of feature parameters to determine the severity index of diabetes disease for the individual under test by matching the training feature parameters with the probe feature parameters.

[0077] In another embodiment the processor 406 may further be configured to match the first set of feature parameters with the second set of feature parameters to determine the progression risk of the diabetes disease for the individual under test.

[0078] In another embodiment the disclosed invention may collect, record, acquire and other physiological parameters using invasive or imaging technique like blood glucose level or echocardiogram respectively and transmit it to central server or cloud or remote device. The processor 406 may further be configured to fuse the diabetes information obtained by implementing the current invention with other physiological parameters to estimate a cardiac risk to the individual under test.