Patent Application
20210212626
This U.S. patent application claims priority under 35 U.S.C. § 119 to: India Application No. 201921051884, filed on Dec. 13, 2019. The entire contents of the aforementioned application are incorporated herein by reference.
The embodiments herein generally relate to the field of cardiac health monitoring. More particularly, but not specifically, the present disclosure provides a device and method to continuously monitor and determine the cardiac health of a person by capturing electrocardiogram (ECG).
Electrocardiogram (ECG) system has been adopted for almost a century to diagnose cardiovascular disease (CVD). Further, monitoring the cardiac signal continuous will provide an insight of CVD and function as an aiding tool for physician towards early detection of cardiac events. For an instant, monitoring the electrocardiogram (ECG) signal of an individual continuously during the day-to-day activities could detect the cardiac arrhythmia like AF during the occurrence. Currently, there is no such direct wrist-worn cardiac solution that could monitor and alert during movement, activity and, other day to day activities.
Currently used method to characterize an individual's cardiac disorder is established by collecting Electrocardiogram (ECG) in a controlled environment or conducting a voluntary study such as treadmill test, Holter system etc. Alternative methods were also adapted during an individual's movement by using PPG or voluntary contribute by touching a pair of electrodes attached either in wrist band like apple watch or phone case like AliveCore or other means in the wearable market. These means of acquiring ECG signal would either interpret or distract once activity and there is a high chance one could ignore its notification, alert, or time for contribution.
In general, none of the prior art have had addressed the acquiring, detection of a cardiac event in real-time or during the occurrence of cardiac disorder event using a wearable device. Besides, all the existing system exist require a voluntary contribution from the human to capture the ECG.
Embodiments of the present disclosure present technological improvements as solutions to one or more of the above-mentioned technical problems recognized by the inventors in conventional systems. For example, in one embodiment, a wearable device for continuous monitoring of cardiac health of a person is provided. The device comprises a first electrode, a second electrode, one or more hardware processors and a memory. The memory further comprises a classifier generation module, and the cardiac health monitoring module. The first electrode either of a contact type or a non-contact type of electrode. The second electrode of non-contact type of electrode, wherein the first electrode and the second electrode are configured to acquire an ECG signal. The classifier generation module further configured to generate a classifier and the classifier is pre-generated. The cardiac health monitoring module further configured to perform the steps of: capturing an ECG signal of the person using the wearable device, preprocessing the captured ECG signal of the person, extracting a plurality of test features from the preprocessed ECG signal, and detecting the presence of the cardiac disorder in the person using the plurality of test features and the classifier.
In another aspect, the embodiment here provides a method for continuous monitoring of cardiac health of a person. Initially, a wearable device is provided, the wearable device comprises a first electrode either of a contact type or a non-contact type and a second electrode of non-contact type, wherein the first electrode and the second electrode are configured to acquire an ECG signal, wherein the wearable device comprising a classifier and the classifier is pre-generated. Further, an ECG signal of the person is captured using the wearable device. In the next step the acquired ECG signal of the person is preprocessed. Further a plurality of test features are extracted from the preprocessed ECG signal. And finally, the presence of the cardiac disorder in the person is detected using the plurality of test features and the classifier.
In another aspect the embodiment here provides one or more non-transitory machine readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause continuous monitoring of cardiac health of a person. Initially, a wearable device is provided, the wearable device comprises a first electrode either of a contact type or a non-contact type and a second electrode of non-contact type, wherein the first electrode and the second electrode are configured to acquire an ECG signal, wherein the wearable device comprising a classifier and the classifier is pre-generated. Further, an ECG signal of the person is captured using the wearable device. In the next step the acquired ECG signal of the person is preprocessed. Further a plurality of test features are extracted from the preprocessed ECG signal. And finally, the presence of the cardiac disorder in the person is detected using the plurality of test features and the classifier.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the disclosed principles.
Exemplary embodiments are described with reference to the accompanying drawings. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. Wherever convenient, the same reference numbers are used throughout the drawings to refer to the same or like parts. While examples and features of disclosed principles are described herein, modifications, adaptations, and other implementations are possible without departing from the scope of the disclosed embodiments.
Referring now to the drawings, and more particularly to
According to an embodiment of the disclosure, a device 100 for continuous monitoring of cardiac health of a person is shown in the schematic overview of
According to an embodiment of the disclosure, the device 100 comprises a first electrode 102, a second electrode 104, a memory 106 and one or more hardware processors 108 as shown in the block diagram of
According to an embodiment of the disclosure, the wearable device 100 utilizes a non-contact sensing mechanism. This could be achieved by using a hybrid sensing technique. Normally, the ECG requires a pair of electrode/sensor of which, one sensor could be in contact sensor say resistive sensor that will be touching the skin and second sensor could be a non-contact sensor say a capacitive based sensor. In an embodiment of the present disclosure, the first electrode 102 is either a contact type or a non-contact type of electrode. The second electrode 104 is a non-contact type of electrode. The first electrode 102 and the second electrode 104 are configured to acquire the ECG signal, wherein the ECG signal of an individual is acquired when the wearable sensor device comes to a predefined position on the body of the individual.
According to an embodiment of the disclosure, the wearable device 100 is worn on the wrist of the person as shown in
In another embodiment of the disclosure, the wearable device 100 can be worn on the neck of the person. The design of the device may vary depending on the place where the person is wearing the wearable device 100. In case of neck wearable device, the first and the second electrode are both the non-contact type of electrodes and can be worn as a necklace. In this case, the device 100 is configured to capture single-lead ECG configuration.
According to an embodiment of the disclosure, the pair of electrode is made of two different material and sensing are performed using two different phenomena. To be more specific, the contact type of electrode senses using skin resistance. The contact type of electrode may be made of sliver or sliver-sliver-chloride. The non-contact electrode will be using capacitive phenomena. The non-contact electrode may be made of copper, gold or platinum.
According to an embodiment of the disclosure, the memory 106 further comprises the classifier generation module 110 as shown in the block diagram of
As shown in the flow diagram of
According to an embodiment of the disclosure, the plurality of features is extracted using the slope based feature extraction method. In this method, the slope of the ECG signal within a window size of ‘n’ number of samples is performed to extract the morphological details. The slope of the ECG signal has both positive and negative values due to increasing and decreasing peaks in an ECG waveform. The slope of the signal is calculated using Equation (1).
S
slope(i)=tan−1(S(i+n)—S(i))/n (1)
where =1, 2 . . . N-n,
The window size depends on the number of samples between the Q peak and R peak in the ECG signal. A standard range of values is defined for the inclination angle of the P wave, QRS complex and T wave for both normal and abnormal ECG. Thus, from the defined range of slope values for the ECG waveform, the slope values between the minimum positive slope value and the maximum negative slope values are removed to eliminate any noise. For finding the window size, the R peak is found by differentiating the ECG signal and the Q wave is detected as the negative peak immediately before the detected R peak. The slope of the signal within this window is found for the entire signal as shown in
The first positive peak is the P_on, the first negative peak is P and the following zero crossings is P_off. Similarly, the procedure has been performed to identify the QRS complex and T wave. The features extracted using the slope method with a mark on the signal plot is shown in
According to an embodiment of the disclosure, the memory 106 further comprises the cardiac health monitoring module 112. The cardiac health monitoring module 112 is configured to acquire the ECG signal of the person captured using the wearable device 100, preprocess the acquired ECG signal of the person and extract a plurality of test features from the preprocessed ECG signal as shown in the flow diagram of
In operation, a flowchart 200 illustrating a method for continuous monitoring of cardiac health of the person as shown in
At step 204, the ECG signal of the person who is being monitored is captured using the wearable device 104. The ECG signal is captured in real-time continuously. At step 206, the acquired ECG signal of the person is preprocessed. At step 208, a plurality of test features is extracted from the preprocessed ECG signal. And finally at step 210, the presence of the cardiac disorder in the person detected using the plurality of test features and the generated classifier. It should be appreciated that the presence of cardiac disorder can be communicated to a distant location. So that a reactive measure can be taken to improve the cardiac health of the person.
According to an embodiment of the disclosure, the wearable device 100 further configured to perform auto impedance mismatching correction.
According to an embodiment of the disclosure, the wearable device 100 further comprises an accelerometer 114 to capture the movement of the individual as shown in
According to an embodiment of the disclosure, the plurality of features such as morphological features extracted was used to detect the arrhythmias. In an example, three different abnormalities have been considered, namely, Sinus bradycardia, Sinus tachycardia and premature ventricular contraction (PVC). Feature parameter were P wave duration, QRS complex durations, T wave duration, PR intervals, QT intervals, and ST intervals are chosen as feature vectors. In an example, two classifiers are used to detect arrhythmias, namely Dynamic time warping (DTW) and Adaboost classifier and their performance were compared. It should be appreciated that any such machine learning or classifier technique could be used by a person skilled in the art
The DTW classifier is based on the ranking of the prototypes by the distance to the query. Let, F=(f1 . . . fn) and G=(g1 . . . gm) be two time series of length n and m, respectively. To align the two sequences using DTW, an n-by-m matrix was constructed whose (i, j) th element is the Euclidean distance d (i, j) between two points fi and gi. The (i, j) th matrix element corresponds to the alignment between the points fi and gi. A warping path, R is a contiguous sets of matrix elements that defines a mapping between F and G and is written as R={r1 . . . rs} where, max (m, n)<S<m+n−1. The warping path is typically subject to several constraints such as boundary conditions, continuity, monotonicity, and windowing. The DTW algorithm finds the point-to-point correspondence between the curves, which satisfies the above constraints and yields the minimum sum of the costs associated with the matching of the data points. There are exponentially many warping paths that satisfy the above conditions. The path that minimizes the warping cost is shown in equation (2),
D(F,G)=minΣs=0srs (2)
The warping path can be found efficiently using dynamic programming to evaluate a recurrence relation, which defines the cumulative distance γ(i, j) up to the element (i, j) as the sum of d (i, j), the cost of dissimilarity between the ith and the jth points of the two sequences and the minimum of the cumulative distances up to the adjacent elements as shown in equation (3):
γ(i,j)=d(i,j)+min{γ(i−1,j),γ(i,j−i),γ(i−1,j−1)} (3)
In an example, the classification procedure based on DTW yielded the following results.
In this study, multiclass AdaBoost has been used in identifying arrhythmia detection. Adaboost classifier increases the accuracy of the weak classifier by reinforcing training on misclassified samples and assigns appropriate weights to each weak classifier. The final classification is given by equation (4)
where 1 indicates that the sample has been correctly classified. In this experiment, stumps are used as a weak classifier. For reassigning the weights to the weak classifier 5000 iterations were performed and this was experimentally found to yield better results.
Because it may have potential advantages such as higher classification performance, more rapid recognition process time and extension of recognition features, Adaboost was applied for the detection of cardiac arrhythmia. Each class of ECG type i.e. normal or arrhythmic, a label +1 or −1 is assigned to it. A large number of weak classifiers around 5000 are chosen. Decision stumps are chosen for classification. Decision stumps make a prediction based on the value of just a single input feature. The input value if greater than the prediction value then the feature vector belongs to one class else it belongs to another class. Initially, a set of training vectors are fed for classification. Labels are assigned for each input. A set of testing vectors are given as inputs for classification. Based on the labels assigned to each of the testing vectors, the classification or misclassification is decided.
The Adaboost classifier is implemented and the classification results are as shown in Table 4. The sensitivity of the classifier is evaluated and the average sensitivity is found to be 99%. Table 3 presents the performance of the classification system for different arrhythmias. The performance of the detection of an arrhythmia is measured on the parameters of false rejection (FR), false acceptance (FA), false acceptance rate (FAR) and false rejection rate (FRR) reported by the system.
The illustrated steps are set out to explain the exemplary embodiments shown, and it should be anticipated that ongoing technological development will change the manner in which particular functions are performed. These examples are presented herein for purposes of illustration, and not limitation. Further, the boundaries of the functional building blocks have been arbitrarily defined herein for the convenience of the description. Alternative boundaries can be defined so long as the specified functions and relationships thereof are appropriately performed. Alternatives (including equivalents, extensions, variations, deviations, etc., of those described herein) will be apparent to persons skilled in the relevant art(s) based on the teachings contained herein. Such alternatives fall within the scope of the disclosed embodiments. Also, the words “comprising,” “having,” “containing,” and “including,” and other similar forms 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. 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.
Furthermore, one or more computer-readable storage media may be utilized in implementing embodiments consistent with the present disclosure. 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., be non-transitory. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, nonvolatile memory, hard drives, CD ROMs, DVDs, flash drives, disks, and any other known physical storage media.
It is intended that the disclosure and examples be considered as exemplary only, with a true scope of disclosed embodiments being indicated by the following claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 201921051884 | Dec 2019 | IN | national |
