DESC:A DEVICE, A SYSTEM AND A METHOD OF NON-INVASIVE DETECTION AND MEASUREMENT OF VITAL SIGNS OF A SUBJECT

TECHNICAL FIELD
The present disclosure in general relates to a field of health monitoring. More particularly, the present disclosure relates to a system, wearable device, electronic device and method for monitoring metabolic health parameter.

BACKGROUND
In recent trend of life, health awareness has become of prime importance. Early detection of any aberration in the vital signs helps to prevent sudden deterioration of a health condition of the human being. This also helps to provide required medical care. For example, monitoring blood glucose levels is crucial for individuals with diabetes and other metabolic disorders.
Traditional methods of glucose monitoring rely on invasive techniques such as finger-prick tests or continuous glucose monitoring (CGM) systems that require subcutaneous sensors. While effective, these approaches are often associated with discomfort, inconvenience, and compliance issues. Non-invasive methods for glucose monitoring, though more user-friendly, have historically faced challenges in achieving the accuracy required for clinical or personal health management applications.
Another significant drawback of the non-CGM device is its inability to provide continuous, and real-time data. Instead, the non-CGM device only offers intermittent measurements of glucose levels at specific time points, such as before or after meals, or as required for monitoring purposes.
As discussed above, one of the other major drawbacks of the invasive methods is that they cause discomfort and occasional pain associated with the prick on the finger or a sensor insertion. The need for a subcutaneous sensor causes inconvenience and, in some cases, deter individuals from consistent use. Further, the insertion of sensors beneath the skin introduces a potential risk of infection. The puncture site becomes a susceptible entry point for bacteria, increasing the likelihood of localized infections and requiring meticulous hygiene practices. Furthermore, the invasive CGM devices often have a limited wear time before the sensor needs replacement. This necessitates frequent insertions, which can be inconvenient and may contribute to increased costs for users.
Therefore, it would be advantageous to have a system that may provide a solution to the aforementioned problems

SUMMARY
Before the present system, wearable device, electronic device and method for monitoring metabolic health parameter is described, it is to be understood that this application is not limited to the particular system, wearable device, electronic device and methodologies described, as there can be multiple possible embodiments that are not expressly illustrated in the present disclosure. It is also to be understood that the terminology used in the description is to describe the particular versions or embodiments only, and is not intended to limit the scope of the present application. This summary is not intended to identify essential features of the claimed subject matter nor is it intended for use in determining or limiting the scope of the claimed subject matter.
In one implementation, a system for monitoring metabolic health parameter is disclosed. The system comprises of a wearable device that comprises of a sensor assembly configured transmit light on a user’s skin to obtain sensor data in real-time. The sensor data is obtained from a reflected signal received in response to the transmitted light. The wearable device further comprises processing circuitry arranged to map the sensor data to one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model and generate metabolic health pattern of the user based on the mapping. The system further comprises an electronic device communicatively coupled to the wearable device configured to receive the metabolic health pattern of the user and generate metabolic health trend based on the metabolic health pattern of the user.
In one implementation, a wearable device for monitoring metabolic health parameter is disclosed. The wearable device comprises of a sensor assembly configured transmit light on a user’s skin to obtain sensor data in real-time. The sensor data is obtained from a reflected signal received in response to the transmitted light. The wearable device further comprises processing circuitry arranged to map the sensor data to one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model and generate metabolic health pattern of the user based on the mapping.
In another implementation, an electronic device for monitoring metabolic health parameter is disclosed. The electronic device comprises of a user interface configured to receive one or more user inputs for the metabolic health parameter at one or more predefined time intervals. The electronic device further comprises of a memory configured to store an application to establish communication with the system and the wearable device and a processor configured to execute one or more instruction stored in the memory. The processor is configured to transmit the one or more user inputs to the wearable device at the one or more predefined time intervals, receive metabolic health pattern from the wearable device and generate a metabolic health trend from the metabolic health pattern received from the wearable device.
In another implementation, a method for monitoring metabolic health parameter is disclosed. The method comprises of emitting, through a sensor assembly of a wearable device, light on a user’s skin to obtain sensor data in real-time. The sensor data is obtained from a reflected signal received in response to the transmitted light. The method further comprises mapping, through the processing circuitry of the wearable device, the sensor data to one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model, generating, through the processing circuitry, metabolic health pattern of the user based on the mapping and transmitting, to an electronic device, the metabolic health pattern of the user.
In another implementation, a method for monitoring metabolic health parameter is disclosed. The method comprises of emitting, through a sensor assembly of a wearable device, light on a user’s skin to obtain sensor data in real-time. The sensor data is obtained from a reflected signal received in response to the transmitted light. The method further comprises mapping, through the processing circuitry of the wearable device, the sensor data to one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model, generating, through the processing circuitry, metabolic health pattern of the user based on the mapping.
In another implementation, a method for monitoring metabolic health parameter is disclosed. The method comprises of receiving, through a user interface, one or more user inputs for the metabolic health parameter at one or more predefined time intervals, establishing, through an application stored in a memory, a communication with a system and a wearable device, transmitting, through processor, the one or more user inputs to the wearable device at the one or more predefined time intervals, receiving, through the processor, metabolic health pattern from the wearable device and generating, through processor, a metabolic health trend from the metabolic health pattern received from the wearable device.
The preceding is a simplified summary to provide an understanding of some embodiments of the present invention. This summary is neither an extensive nor exhaustive overview of the present invention and its various embodiments. The summary presents selected concepts of the embodiments of the present invention in a simplified form as an introduction to the more detailed description presented below. As will be appreciated, other embodiments of the present invention are possible utilizing, alone or in combination, one or more of the features set forth above or described in detail below.

BRIEF DESCRIPTION OF DRAWINGS
The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identify the figure in which the reference number first appears. The same numbers are used throughout the drawings to refer like features and components.
FIG. 1 illustrates a network implementation diagram (1000) of a system (102), wearable device (100) and an electronic device (200) for for monitoring metabolic health parameter, in accordance with an embodiment of the present subject matter;
FIG. 2A illustrates a block diagram of the wearable device (100) for monitoring metabolic health parameter, in accordance with an embodiment of the present subject matter;
FIG. 2B illustrates an arrangement of plurality of Light Emitting Diodes (LEDs) (104a, 104b, 104c) in a sensor assembly (102) of the wearable device (100), in accordance with an embodiment of the present subject matter;
FIG. 2C illustrates an exemplary diagram of a bus (116) in the wearable device (100) of FIG. 1 and FIG. 2A, in accordance with an embodiment of the present subject matter;
FIG. 2D illustrates an exemplary use case scenario of the wearable device (100), in accordance with an embodiment of the present subject matter;
FIG. 3 illustrates a block diagram of the electronic device (200), in accordance with an embodiment of the present subject matter;
FIG. 4 illustrates a flow diagram of a method (400) for monitoring metabolic health parameter, in accordance with an embodiment of the present subject matter;
FIG. 5 illustrates a flow diagram of a method (500) for monitoring metabolic health parameter, in accordance with an alternate embodiment of the present subject matter; and
FIG. 6 illustrates a flow diagram of a method (500) for monitoring metabolic health parameter, in accordance with another alternate embodiment of the present subject matter.
DETAILED DESCRIPTION

Some embodiments of the present disclosure, illustrating all its features, will now be discussed in detail. The words “comprising”, “receiving”, “determining”, “assigning” 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. 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 disclosure, the exemplary, systems and methods for recommending personalized makeup are now described. The disclosed embodiments of the systems and methods recommending personalized makeup are merely exemplary of the disclosure, which may be embodied in various forms.
Various modifications to the embodiment will be readily apparent to those skilled in the art and the generic principles herein may be applied to other embodiments. However, one of ordinary skill in the art will readily recognize that the present disclosure for recommending personalized makeup is not intended to be limited to the embodiments illustrated but is to be accorded the widest scope consistent with the principles and features described herein.
The present subject matter overcomes the problems involved in the existing system and method for monitoring metabolic health parameter, for example blood glucose levels. Proposed solution comprises a system, a wearable device, an electronic device and a method for monitoring metabolic health parameter.
Referring now to the drawings, and more particularly to FIGs 1 through 6, where similar reference characters denote corresponding features consistently throughout the figures, there are shown embodiments.
In accordance with an embodiment, referring to FIG. 1, a network implementation 1000 of a system 1002 for monitoring metabolic health parameter is disclosed. In one example, the system 1002 may be connected with one or more wearable device(s) 100, one or more electronic device(s) 200 and other devices through a communication network 1004.
It should be understood that the system 1002, the wearable device 100, and the electronic device 200 (alternatively may referred as user devices) correspond to computing devices. It may be understood that the system 1002 may also be implemented in a variety of computing systems, such as a laptop computer, a desktop computer, a notebook, a workstation, a mainframe computer, a server, a network server, a cloud-based computing environment, or a smart phone and the like. It may be understood that the mobile devices 104 may correspond to a variety of a variety of portable computing devices, such as a laptop computer, a desktop computer, a notebook, a smart phone, a tablet, a phablet, and the like.
In one implementation, the communication network 1004 may be a wireless network, a wired network, or a combination thereof. The communication network 1004 may be implemented as one of the different types of networks, such as intranet, Local Area Network (LAN), Wireless Personal Area Network (WPAN), Wireless Local Area Network (WLAN), wide area network (WAN), the internet, and the like. The communication network 1004 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, MQ Telemetry Transport (MQTT), Extensible Messaging and Presence Protocol (XMPP), 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 communication network 1004 may include a variety of network devices, including routers, bridges, servers, computing devices, storage devices, and the like.
In accordance with an embodiment, referring to FIG. 2, a block diagram of the wearable device 100 is shown.
The wearable device 100 may comprise a sensor assembly 102 and at least one processing circuitry 104. The wearable device further comprises a Printed Circuitry board 108, an interface 110, a display 112, a battery 114, a bus 116 and a transceiver 118.
The processing circuitry 104 may comprise one or more commonly known Central Processing Units (CPUs) such as a microprocessor or microcontroller. It should be understood that the processing circuitry 104 might be responsible for implementing specific functions under the control of software including an operating system, and any appropriate applications software. In one example, the processing circuitry 104 is configured or designed to store data, program instructions. The program instructions might control the operation of an operating system and/or one or more applications. The processing circuitry 104 is housed on a bus or bus bar 116. In some implementations, the wearable device 100 includes a memory (not shown) configured for storing the program instructions and data processed by the processing circuitry 104.
The interface 110 may include wired interfaces and/or wireless interfaces. In at least one implementation, the interface(s) 110 may include functionality similar to at least a portion of functionality implemented by one or more computer system interfaces such as those described herein and/or generally known to one having ordinary skill in the art.
The display 112 may be implemented using LCD display technology, OLED display technology, and/or other types of conventional display technology. The display 112 configures to display text or images processed by the processing circuitry 104.
The wearable device 100 may further comprise the battery 114. The battery 114 includes a rechargeable battery such as a lithium-ion, for example. In one example, the battery 114 is configured to charge using a wired or wireless power source as known in the art.
The wearable device 100 may include a wireless communication module/transceiver 118. The transceiver 118 may be configured to communicate with external devices using one or more wireless interfaces/protocols such as, for example, 802.11 (Wi-Fi), 802.15 (including Bluetooth™), 802.15 (Wi-Max), 802.22, Cellular standards such as CDMA, CDMA2000, WCDMA, Radio Frequency (e.g., RFID), Infrared, Near Field Magnetics, etc.
In an example, as shown in FIG. 2D later, the wearable device 100 comprises a band designed using medical-grade silicon, ensuring comfort and durability for the user.
In accordance with an embodiment, referring to FIG. 2B and FIG. 2C in combination, the sensor assembly 102 is configured to transmit light on a user’s skin to obtain sensor data in real-time. The sensor data is obtained from a reflected signal received in response to the transmitted light.
In an example, the sensor data comprises characteristics of the reflected signal. The characteristics comprises an amplitude of the reflected signal, frequency components associated with the reflected signal, phase information of the reflected signal, waveform shape of the reflected signal, a signal slope of the reflected signal, optical absorption characteristics of the reflected signal, baseline shifts and trends of the reflected signal, and temporal variations in the reflected signal.
In an example, the amplitude comprises a strength or intensity of the light reflected or absorbed, indicating variations in tissue properties like blood flow or glucose levels of the user. The frequency components comprise spectral features derived from the reflected signal using techniques like Fast Fourier Transform (FFT) to identify periodic variations corresponding to physiological states of the user. The phase information comprises timing or phase shifts in the reflected signal relative to transmitted light, which may provide additional data about the absorption and reflection pattern. The waveform shape comprises an overall shape or morphology of the reflected signal over time, such as peaks, valleys, or transitions, which might correlate with metabolic health parameter. The signal slope comprises a rate at which the reflected signal rises or falls, which may help identify dynamic changes in blood flow or glucose-related absorption. The optical absorption characteristics may be derived from the wavelengths of the plurality of LEDs 104a, 104b, 104c used and indicate how much light is absorbed by glucose or other constituents in the tissue. The baseline shifts and trends provide changes in the reflected signal baseline over time that may indicate overall metabolic trends or shifts in user health states. The temporal variations comprise short-term or long-term fluctuations in the signal corresponding to events like fasting, physical activity, or glucose intake.
The sensor assembly 102 comprises a plurality of Light Emitting Diodes (LEDs) 104a, 104b, 104c configured to emit light onto the user’s skin. One or more LEDs 104b, 104c of the plurality of LEDs 104a, 104b, 104c are arranged diagonally to one or more other LEDs 104a of the plurality of LEDs 104a, 104b, 104c. The plurality of LEDs 104a, 104b, 104c comprise one or more infrared (IR) LED 104a emitting light selected in a wavelength range of 940–1200 nm, red LEDs 104b or green LEDs 104c. The red LEDs 104b and the green LEDs 104c are arranged diagonally with respect to the IR LED 104a for transmitting the light onto the user’s skin.
The wavelength range of 940–1200 nm for the IR LED 104a is particularly effective for applications requiring non-invasive measurements and low-light conditions. This wavelength is optimal for minimizing interference from ambient light, ensuring accurate readings of the IR LED 104a. The configuration of the green LED 104c and the red LEDs 104b enhances the sensor assembly’s 102 versatility. By placing IR LED 104a diagonally to red LED 104b and green LED 104c (opposite each other), the sensor assembly 102 provides a balanced light emission that may effectively penetrate various surfaces, making the sensor assembly 102 suitable for a wide range of materials and conditions.
In an example, in the proposed wearable device 100, the alignment of the IR LED 104a with the red LEDs 104b and the green LEDs 104c is designed to facilitate improved data correlation, which is critical for effective analysis in glucose monitoring applications of the wearable device 100. By integrating the sensor assembly 102 with advanced AI algorithms and adaptive prediction model (machine learning models), the wearable device 100 may process the sensor data and the one or more user inputs for deeper insights and more accurate glucose level predictions. This capability of the wearable device 100 not only enhances real-time monitoring but also allows for personalized health recommendations based on the analysed sensor data in real-time, significantly advancing the state of non-invasive glucose monitoring technologies.
The diagonal placement of the plurality of LEDs 104a, 104b, 104c promotes better light transmission and reduces shadowing effects. The diagonal arrangement of the plurality of LEDs 104a, 104b, 104c allows for a more uniform light field, enhancing the ability of the sensor assembly 102 to detect subtle variations in light intensity and improving the overall accuracy of measurements.
The sensor assembly 102 further comprises a central photodiode 104d configured to detect a light reflected or absorbed by the skin. The light reflected is used to obtain the reflected signal used to obtain the metabolic health pattern of the user. The central photodiode 104d is high-sensitivity photodiode that is positioned centrally in the sensor assembly 102. This design of the sensor assembly 102 maximizes detection area in the skin, allowing for precise measurement of the light reflected back from a target measurement. The central photodiode 104d is engineered to respond effectively to all three wavelengths emitted by the plurality of LEDs 104a, 104b and 104c, ensuring accurate readings regardless of the light source.
The PCB 108 is configured for interconnecting the plurality of LEDs 104a, 104b, 104c and the central photodiode 104d.
The sensor assembly 102 may be configured for various applications, including object detection, proximity sensing, and environmental monitoring. The diagonal arrangement of the plurality of LEDs 104a, 104b, 104c is a configuration makes the sensor assembly 102 particularly well-suited for use in robotics, automation, and smart home devices.
FIG. 2C shows an exemplary architectural diagram of the bus 116 housing the processing circuitry 104, first sensor 104a and second sensors 104b, 104c. As may be seen, the second sensors 104b, 104c include four IR sensors positioned around the first sensor 104a i.e., central sensor. In the present subject matter, the second sensors 104b, 104c are configured to operate in 940nm wavelength to detect infrared radiation in their environments and output electric signals.
Details of the wearable device 100 will now be explained. The wearable device 100 is configured to monitor the metabolic health parameter by predicting the metabolic health parameter at one or more predefined time intervals. In an example, the metabolic health parameter comprises blood glucose levels (also referred as glucose or sugar levels).
The wearable device 100 is configured to receive one or more user inputs (manual inputs for actual Blood Glucose Level or BGL data) for the metabolic health parameter at one or more predefined time intervals. The one or more predefined time intervals comprise a fasting time interval to record metabolic health parameter, and one or more post prandial time intervals (i.e., after breakfast, after lunch or after dinner time intervals) to record the metabolic health parameter. The one or more user inputs may be received over a predefined periodic time, for example, the one or more user inputs may be received every 90 days after which the user may re-enter the one or more user inputs.
The processing circuitry 104 of the wearable device 100 is configured to compare the one or more user inputs for the metabolic health parameter with the sensor data using one or more AI algorithms to obtain a correlation between the one or more user inputs and the sensor data and calibrate the wearable device 100 based on the correlation. The calibration is performed to obtain one or more prestored threshold values (obtained and stored in the wearable device 100) of the metabolic health parameter for predicting the metabolic health parameter of the user.
In an exemplary embodiment, once the actual BGL data has been collected for 3 days, the processing circuitry 104 may execute one or more Artificial Intelligence (AI) algorithm to start mapping of the actual BGL values to the sensor data. The AI algorithm may first compare the BGL data with the sensor data and may also use historical data and past calibration results from other users (when available) to fine-tune the mapping of the BGL data and the sensor data. The processing circuitry 104 may then identify patterns and correlations between the BGL data (specific to the user’s demographics) and the sensor data.
After the mapping, the processing circuitry 104 uses AI algorithm to generate a calibration curve that adjusts the sensor data to better reflect or correct the BGL data. The calibration enhances accuracy of the wearable device 100 in predicting real-time metabolic health parameter i.e., the BGL levels or values.
In an example, the measurement of the metabolic health parameter through the wearable device 100 calibrated through the AI algorithms may be valid for 90 days. During this period of 90 days, the sensor assembly 102 may rely on the calibrated wearable device 100 to deliver accurate BGL readings.
After 90 days, the calibration of the wearable device 100 may expire. Users may be notified to re-enter new BGL data for another 3-day period, repeating the calibration process for the wearable device 100. The periodic calibration process may ensure that the sensor assembly 102 remains accurate over time, accounting for any biological changes in the user (e.g., health status, medication changes, etc.). As the AI algorithms accumulates more BGL data and sensor data across multiple cycles (past data from the user and across a broader user base), the wearable device 100 may refine mapping algorithms used for mapping the one or more user inputs with the sensor data leading to progressively higher accuracy of the wearable device 100.
In an example, the use of the one or more AI algorithms for calibration of the wearable device 100 will now be discussed.
Step 1: Input Data Pipeline (actual BGL data received from the one or more users). The data pipeline is structured to ensure robust and efficient handling of raw sensor data.
Step 2: Raw sensor data acquisition from the sensor assembly 102:
Infrared (IR) sensor 104a, 104b, 104c outputs spectral signals capturing skin tissue absorption and reflection patterns.
Each sample is a time-series vector X = {x_1, x_2, ..., x_n}, where (x_i ) represents filtered sensor values.
Step 3: Signal Preprocessing by the processing circuitry 104:
A custom filtering mechanism removes environmental noise
Adaptive Band pass Filter: Retains the IR signal range associated with glucose absorption.
Noise Reduction: Applies a median filter to eliminate sudden spikes caused by external interference.
Normalization: Sensor data are normalized to a [0, 1] range using dynamic baselines calculated per user.
Step 4: Calibration of the wearable device 100:
During the initial three-day user calibration period:
User Input Integration:
Manually entered glucose values G = {g_1, g_2, ..., g_t} \) are paired with the corresponding sensor data vectors \X = {X_1, X_2, ..., X_t} .
Baseline Mapping:
A regression model M_calib is trained locally.
M_calib (X_i )=g_i,?_i?[1,t]

The model identifies a user-specific mapping function (f_ map) that correlates normalized sensor features to blood glucose levels.

Feature Engineering:

Once calibration of the wireless device 100 is complete:
Derived Features:
Spectral Signal Derivatives: First and second-order derivatives of ( X ) are computed to detect minute fluctuations.
Temporal Aggregates: Rolling mean and variance of sensor values over predefined windows (e.g., 5 samples) capture temporal trends.
Fourier Transform Features: Extract frequency-domain characteristics of IR signals, highlighting glucose-related spectral peaks.
Dimensionality Reduction:
Principal Component Analysis (PCA) is applied to reduce redundant features, optimizing computational efficiency for deployment.
Prediction Model Architecture (adaptive prediction model):
The prediction model is structured as follows:
Input Layer: Accepts the processed feature vector F = {f_1, f_2, ..., f_m} , where ( m ) is the dimensionality after feature engineering.
Hidden Layers:
Layer 1:
Neurons: 64
Activation: ReLU
Batch Normalization: Ensures stable training.
Layer 2:
Neurons: 32
Activation: ReLU
Dropout: 20% for regularization.
Output Layer:
Single neuron with linear activation for continuous glucose level prediction (prediction of the metabolic health parameter)
g ^= M_predict (F)

Training Objective for training the adaptive prediction model:
Loss Function: Mean Absolute Percentage Error (MAPE), ensuring error minimization across diverse glucose ranges.
MAPE= 1/N ?_(i=1)^N¦?|((g_i ) ^-g ^_i)/g_i |×100?
Adaptation and Personalization:
Model Update (updating the adaptive prediction model):
Periodic retraining is triggered upon user feedback (e.g., manual blood test values)
A lightweight on-device retraining algorithm incorporates new data points while retaining past knowledge using a modified Stochastic Gradient Descent (SGD) optimizer.
Adaptive Regularization:
Elastic Net Regularization (L1 + L2 penalties) prevents overfitting during updates
Loss=MAPE+?_1 ?¦?|?|+ ? ?_2 ?¦??^2 ?

Deployment Algorithm Workflow
Real-Time Prediction Workflow
Acquire raw sensor data X_real-time.
Preprocess using the noise filtering and normalization pipeline.
Transform into feature vector F_real-time.
Predict glucose level:
g ^_(real-time)= M_predict (F_(real-time) )
Display result on the wearable device interface
Continuous Calibration of the wearable device 100
User inputs occasional blood test results.
Fine-tune the adaptive prediction model
Incorporate new data points (X_new, g_new).
Retrain using cumulative past data to refine the mapping function of the adaptive prediction model
In an example, once the wearable device 100 is calibrated (as discussed above) and the adaptive prediction model is trained, the processing circuitry 104 receives the sensor data (raw sensor data) to map the sensor data to the one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model to generate metabolic health pattern of the user based on the mapping. The one or more user inputs and the sensor data is used to train the adaptive prediction model (machine learning model).
The processing circuitry 104 is configured for identifying a relationship between characteristics of the reflected signal of the sensor data with the one or more prestored threshold values using the adaptive prediction model and map the characteristics of the reflected signal with the one or more prestored threshold values based on the identification. The one or more prestored threshold values represent predefined ranges of metabolic health parameters. The predefined ranges comprise normal range or abnormal range for the metabolic health parameter.
In an example, the mapping comprises feature matching for the senor data with the one or more user inputs. The characteristics of the reflected signal from the sensor output (e.g., amplitude, frequency, or absorption coefficients) are compared to corresponding reference values in the threshold database (prestored threshold values). The adaptive prediction model uses the one or more AI algorithms to recognize patterns in the sensor data and associate them with specific thresholds that represent glucose ranges or metabolic states. The adaptive prediction model minimizes a difference between predicted values (based on sensor data) and known predefined thresholds by recalibrating the wearable device 100 over time.
The wearable device 100 determines whether the sensor data falls within normal or abnormal thresholds. For example, if the sensor output aligns with thresholds corresponding to 70-99 mg/dL, the wearable device 100 interprets the glucose level as normal, and if the sensor data exceeds an upper limit of the one or more threshold values (e.g., >140 mg/dL postprandial), the wearable device 100 may generate the metabolic health parameter as a high glucose level.
The metabolic health pattern comprises values of the metabolic health parameter at one or more predefined time intervals. The predefined time intervals may comprise morning time to measure fasting glucose levels, or after breakfast, after lunch and after dinner time interval to measure post meal glucose levels. The predefined time interval may also be selected by the user at any point of time.
In an example, the one or more AI algorithms used for training the adaptive prediction model comprises semi-supervised learning for mapping the sensor data with the one or more user inputs, deep neural networks for initial training of the adaptive prediction model, incremental learning for periodic updates on the metabolic health pattern, and feature engineering techniques for processing the sensor data to generate the metabolic health pattern.
The processing circuitry 104 is further configured to process the sensor data using one or more signal preprocessing techniques. The signal processing techniques comprises of filtering technique to filter noise from the sensor data, normalization technique to normalize the sensor data, or dimensionality reduction technique to extract characteristics of the reflected signal as discussed above.
In accordance with an embodiment, referring to FIG. 3, a block diagram for the electronic device 200 for monitoring the metabolic health parameter is shown. The electronic device 200 comprises an input/output interface 202, a processor 204, a memory 206, a display 208, a transceiver 210 and a battery 212.
In an implementation, the electronic device 200 may be registered and paired with the system 1002 and synchronized with the wearable device 100. The electronic device 200 is having an application stored in the memory 206 for establishing communication with the system 1002 and the wearable device 100.
The user of the electronic device 200 may register with the system 1002 (as discussed above) by creating an account on the system 1002 by providing basic demographic information such as age, gender, medical history, and lifestyle factors as user registration data. The user registration data is essential for the calibration of the wearable device 100 based on individual differences. After account creation, the user may log in to the system 1002 using secure credentials of the user. Returning users may access their profiles, historical data, and calibration status of the wearable device 100.
The user may enter the one or more user inputs for the metabolic health parameter at one or more predefined time intervals. The one or more user inputs may be entered via the user interface 202 (input/output interface) of the electronic device 200. The one or more user inputs are of use for sharing with the wearable device 100 for calibration and recalibration of the wearable device 100, as discussed above. The one or more user inputs comprise actual Blood Glucose Level (BGL) data of the user which is used for the calibration of the wearable device 100.
In an example, over a course of 3 days, the user is required to input his/her actual BGL values both during fasting and after food consumption (postprandial values) considered as the one or more predefined time intervals. The fasting BGL values are entered each morning, i.e., after food values are entered at least 2 hours post-meal, twice a day. The actual BGL data may be obtained through lab testing or any other monitoring device.
The transceiver 210 of the electronic device 200 may be prompted by the system 1002 to enter values at regular time intervals.
The calibration of the wearable device 100 based on the one or more user inputs shared by the one or more users of the electronic device 200 is discussed above and hence is not repeated for the sake of brevity.
The processor 202 uses the one or more AI algorithms to analyse metabolic health pattern comprising the glucose level readings and presents the metabolic health pattern to the user in a clear and user-friendly manner. The application is designed to be user-friendly, offering a seamless and intuitive experience. The application in the electronic device 200 provides comprehensive access to all important health parameters, comprising blood glucose levels, heart rate, SPO2, temperature, and blood pressure.
In an example, the electronic device 200 through the application also generates a live visual glucose value chart in the metabolic health pattern, which is then continuously displayed over the display 208 to display changes in glucose readings of the user at different time in real time. The live visual glucose value chart allows users to easily track their glucose levels and detect trends or pattern over time.
The display 208 of the electronic device 200 is configured to allow the users to view and manage their metabolic health parameter effortlessly. The electronic device 200 provides comprehensive health monitoring that comprises displaying one or more metabolic health parameters, including glucose levels, heart rate, SPO2, temperature, and blood pressure. The electronic device 200 is also configured to notify users of any abnormal changes in their metabolic health parameter for timely interventions.
In an embodiment, the electronic device 200 is configured to provide data security. The electronic device 200 uses the application that includes terms and conditions for retrieval of user’s personal information or metabolic health parameter in terms of metabolic health pattern or trend. The retrieval requires an electronic signature (E-Sign) for access, ensuring user privacy and data protection.
The application is equipped with the one or more AI algorithms to process the metabolic health pattern collected by the device from the wearable device 100. The one or more AI algorithms analyse the glucose level readings and presents the required data to the user in a clear and user-friendly manner.
In accordance with an embodiment, referring to FIG. 4, a flow chart for a method 300 for monitoring metabolic health parameter is shown. The method 300 may be executed through the system 1002 as discussed above. The method 300 comprises at step 302, emitting light on the user’s skin to obtain the sensor data in real-time. The emission is performed through the sensor assembly 102 of the wearable device 100. The sensor data is obtained from the reflected signal received in response to the transmitted light.
At step 304, the method 300 provides mapping the sensor data to one or more prestored threshold values of the metabolic health parameter using the adaptive prediction model. The mapping may be performed through the processing circuitry 104 of the wearable device 100. At step 306, the method 300 provides generating the metabolic health pattern of the user based on the mapping. The generating is performed through the processing circuitry 104 of the wearable device 100. At step 308, the method 300 comprises transmitting the metabolic health pattern of the user to the electronic device 200. The transmission is performed through the processing circuitry 104.
Details of the method 300 are similar to the details of the system 1002 as discussed above hence are not repeated for the sake of brevity.
In accordance with an embodiment, referring to FIG. 5, a flow chart for a method 400 for monitoring metabolic health parameter is shown. The method 400 may be executed through the wearable device 100 as discussed above. The method 400 comprises of at step 402, emitting light on the user’s skin to obtain the sensor data in real-time. The emission is performed through the sensor assembly 102 of the wearable device 100. The sensor data is obtained from the reflected signal received in response to the transmitted light.
At step 404, the method 400 provides mapping the sensor data to one or more prestored threshold values of the metabolic health parameter using the adaptive prediction model. The mapping may be performed through the processing circuitry 104 of the wearable device 100. At step 406, the method 400 provides generating the metabolic health pattern of the user based on the mapping. The generating is performed through the processing circuitry 104 of the wearable device 100.
Details of the method 400 are similar to the details of the wearable device 100 as discussed above hence are not repeated for the sake of brevity.
In accordance with an embodiment, FIG. 6 shows flow chart for a method 500 for monitoring metabolic health parameter. The method 500 may be executed by the electronic device 200. The method 500 at step 502 comprises receiving the one or more user inputs for the metabolic health parameter at one or more predefined time intervals. The one or more user inputs are received through the user interface 202. At step 504, the method 500 provides establishing the communication with the system 1002 and the wearable device (100) through the application stored in a memory 202.
The method 500 at step 506 provides transmitting the one or more user inputs to the wearable device 100 at the one or more predefined time intervals through the processor 204. The method 500 at step 508 provides receiving the metabolic health pattern from the wearable device 100 through the processor 204 and at step 510, the method 500 provides generating the metabolic health trend from the metabolic health pattern received from the wearable device 100 through the processor 204.
Details of the method 500 are similar to the details of the electronic device 200 as discussed above hence are not repeated for the sake of brevity.
Exemplary advantages of the proposed system 1002, the wearable device 100, the electronic device 200 and the methods 300, 400 and 500 are discussed below:
The proposed wearable device 100 provides non-invasive monitoring of the metabolic health parameter by eliminating a need for pricking or drawing blood for glucose monitoring, enhancing user convenience and compliance.
The proposed system 1002, the wearable device 100, the electronic device 200 and the methods provide continuous monitoring of the metabolic health parameters, offering real-time insights into glucose trends.
The proposed system 1002, the wearable device 100, the electronic device 200 and the methods adapts to individual user physiology through machine learning, ensuring accuracy tailored to the specific user.
The proposed system 1002, the wearable device 100, the electronic device 200 and the methods combines short-term readings and long-term trends to provide a holistic view of the user’s metabolic health.
The proposed system 1002, the wearable device 100, the electronic device 200 and the methods predict potential health risks using advanced AI algorithms, enabling proactive intervention.
The proposed system 1002, the wearable device 100, the electronic device 200 and the methods uses user inputs and adaptive algorithms for periodic recalibration, ensuring long-term reliability of the predictions.
The wearable device 100 uses a combination of IR, red, and green LEDs with a central photodiode for enhanced accuracy in detecting blood glucose levels.
The proposed system 1002, the wearable device 100, the electronic device 200 and the methods incorporates user-provided glucose levels to fine-tune the adaptive prediction model.
While the invention has been described in connection with a number of embodiments and implementations, the invention is not so limited but covers various obvious modifications and equivalent arrangements, which fall within the purview of the appended claims. Although features of the invention are expressed in certain combinations among the claims, it is contemplated that these features can be arranged in any combination and order.
The exemplary embodiments of this present invention have been described in relation to monitoring of metabolic health parameter. However, to avoid unnecessarily obscuring the present invention, the preceding description omits a number of known structures and devices. This omission is not to be construed as a limitation of the scope of the present invention. Specific details are set forth by use of the embodiments to provide an understanding of the present invention. It should however be appreciated that the present invention may be practiced in a variety of ways beyond the specific embodiments set forth herein.
A number of variations and modifications of the present invention can be used. It would be possible to provide for some features of the present invention without providing others.
The present invention, in various embodiments, configurations, and aspects, includes components, methods, processes, systems and/or apparatus substantially as depicted and described herein, including various embodiments, sub-combinations, and subsets thereof. Those of skill in the art are able to understand how to make and use the present invention after understanding the present disclosure. The present invention, in various embodiments, configurations, and aspects, includes providing devices and processes in the absence of items not depicted and/or described herein or in various embodiments, configurations, or aspects hereof, including in the absence of such items as may have been used in previous devices or processes, e.g., for improving performance, achieving ease and/or reducing cost of implementation.
The foregoing discussion of the present invention has been presented for purposes of illustration and description. It is not intended to limit the present invention to the form or forms disclosed herein. In the foregoing Detailed Description, for example, various features of the present invention are grouped together in one or more embodiments, configurations, or aspects for the purpose of streamlining the disclosure. The features of the embodiments, configurations, or aspects may be combined in alternate embodiments, configurations, or aspects other than those discussed above. This method of disclosure is not to be interpreted as reflecting an intention the present invention requires more features than are expressly recited in each claim. Rather, as the following description reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment, configuration, or aspect. Thus, the following description are hereby incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of the present invention.
,CLAIMS:We claim:
1. A system (1002) for monitoring metabolic health parameter, the system (1002) comprising:
a wearable device (100), comprising:
a sensor assembly (102) configured to:
transmit light on a user’s skin to obtain sensor data in real-time, wherein the sensor data is obtained from a reflected signal received in response to the emitted light;
a processing circuitry (104) arranged to:
map the sensor data to one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model; and
generate metabolic health pattern of the user based on the mapping; and
an electronic device (200) communicatively coupled to the wearable device (100), wherein the electronic device (200) is configured to:
receive the metabolic health pattern of the user; and
generate metabolic health trend based on the metabolic health pattern of the user.

2. The system (1002) as claimed in claim 1, wherein the sensor assembly (102) comprises:
a plurality of Light Emitting Diodes (LEDs) (104a, 104b, 104c) configured to emit light onto the user’s skin, wherein one or more LEDs (104a) of the plurality of LEDs (104a, 104b, 104c) are arranged diagonally to one or more other LEDs (104b, 104c) of the plurality of LEDs (104a, 104b, 104c); and
a central photodiode (104d) configured to detect a light reflected or absorbed by the skin, wherein the light reflected is used to obtain the metabolic health pattern; and
a Printed Circuit Board (PCB) (108) for interconnecting the plurality of LEDs (104a, 104b, 104c) and the central photodiode (104d).

3. The system (1002) as claimed in claim 1, wherein the plurality of LEDs (104a, 104b, 104c) comprise one or more infrared (IR) LED (104a) emitting light selected in a wavelength range of 940–1200 nm, one or more red LEDs (104b) and one or more green LEDs (104c),
wherein the one or more red LEDs (104b) and the one or more green LEDs (104c) are arranged diagonally with respect to the IR LED (104a) for transmitting the light onto the user’s skin.

4. The system (1002) as claimed in claim 1, wherein the processing circuitry (104) is configured to:
process the sensor data using one or more signal pre-processing techniques, wherein the signal pre-processing techniques comprises: filtering technique to filter noise from the sensor data, normalization technique to normalize the sensor data, or dimensionality reduction technique to extract characteristics of the reflected signal.

5. The system (1002) as claimed in claim 1, wherein the sensor data comprises characteristics of the reflected signal, wherein the characteristics comprises an amplitude of the reflected signal, frequency components associated with the reflected signal, phase information of the reflected signal, waveform shape of the reflected signal, a signal slope of the reflected signal, optical absorption characteristics of the reflected signal, baseline shifts and trends of the reflected signal, and temporal variations in the reflected signal.

6. The system (1002) as claimed in claim 1, wherein the processing circuitry (104) is configured to:
identify a relationship between characteristics of the reflected signal of the sensor data with the one or more prestored threshold values using the adaptive prediction model; and
map the characteristics of the reflected signal with the one or more prestored threshold values based on the identification, wherein the one or more prestored threshold values represent predefined ranges of metabolic health parameters, wherein the predefined ranges comprise normal range or abnormal range for the metabolic health parameter.

7. The system (1002) as claimed in claim 1, wherein the metabolic health parameter comprises blood glucose levels, wherein the metabolic health pattern comprises values of the metabolic health parameter at one or more predefined time intervals, and
wherein the health trend comprises normal and abnormal values of the metabolic health parameter with respect to one or more normal values of the metabolic health parameter at the one or more predefined interval and an average of value of the metabolic health parameter over a predefined time period.

8. The system (1002) as claimed in claim 1, wherein the processing circuitry (104) is configured to:
receive, one or more user inputs for the metabolic health parameter at one or more predefined time intervals, wherein the one or more predefined time intervals comprise a fasting time interval to record metabolic health parameter, and one or more post prandial time intervals to record the metabolic health parameter, wherein the one or more user inputs are received over a predefined periodic time;
compare the one or more user inputs for the metabolic health parameter with the sensor data using one or more AI algorithms to identify a correlation between the one or more user inputs and the sensor data; and
calibrate the wearable device (100) based on the correlation.

9. A wearable device (100), comprising:
a sensor assembly (102) configured to:
transmit light on a user’s skin to obtain sensor data in real-time, wherein the sensor data is obtained from a reflected signal received in response to the emitted light; and
a processing circuitry (104) arranged to:
map the sensor data to one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model; and
generate metabolic health pattern of the user based on the mapping.

10. An electronic device (200) for monitoring metabolic health parameter, the electronic device (200) comprising:
a user interface (202) configured to receive one or more user inputs for the metabolic health parameter at one or more predefined time intervals;
a memory (206) configured to store an application to establish communication with the system (1002) and the wearable device (100); and
a processor (204) configured to execute one or more instruction stored in the memory (206), wherein the processor (206) is configured to:
transmit the one or more user inputs to the wearable device (100) at the one or more predefined time intervals;
receive metabolic health pattern from the wearable device (100); and
generate a metabolic health trend from the metabolic health pattern received from the wearable device (100).
11. The electronic device (100) as claimed in claim 10, wherein the metabolic health parameter comprises blood glucose levels, wherein the metabolic health pattern comprises values of the metabolic health parameter at one or more predefined time intervals, and wherein the metabolic health trend comprises normal and abnormal values of the metabolic health parameter with respect to one or more normal values of the metabolic health parameter at the one or more predefined interval and an average of values of the metabolic health parameter over a predefined time period.

12. A method (300) for monitoring metabolic health parameter, the method (300) comprising:
emitting (302), through a sensor assembly (102) of a wearable device (100), light on a user’s skin to obtain sensor data in real-time, wherein the sensor data is obtained from a reflected signal received in response to the emitted light;
mapping (304), through a processing circuitry (104) of the wearable device (100), the sensor data to one or more prestored threshold values of the metabolic health parameter using an adaptive prediction model;
generating (306), through the processing circuitry (104), metabolic health pattern of the user based on the mapping; and
transmitting (308), to an electronic device (200), the metabolic health pattern of the user.

13. The method (300) as claimed in claim 12, comprising:
arranging diagonally, one or more LEDs (104a) of a plurality of LEDs (104a, 104b, 104c) to one or more other LEDs (104b, 104c) of the plurality of LEDs (104a, 104b, 104c);
emitting, through the plurality of Light Emitting Diodes (LEDs) (104a, 104b, 104c), light onto the user’s skin;
detecting, through a central photodiode (104d), a light reflected or absorbed by the skin, wherein the light reflected is used to obtain the metabolic health pattern; and
interconnecting, through a Printed Circuit Board (PCB) (108) the plurality of LEDs (104a, 104b, 104c) and the central photodiode (104d).

14. The method (300) as claimed in claim 12, comprising:
processing, through the processing circuitry (104), the sensor data using one or more signal pre-processing techniques, wherein the signal processing techniques comprises:
filtering technique to filter noise from the sensor data, normalization technique to normalize the sensor data, or dimensionality reduction technique to extract characteristics of the reflected signal.

15. The method (300) as claimed in claim 12, comprising:
identifying, through the processing circuitry (104), a relationship between characteristics of the reflected signal of the sensor data with the one or more prestored threshold values using the adaptive prediction model; and
mapping, through the processing circuitry (104), the characteristics of the reflected signal with the one or more prestored threshold values based on the identification, wherein the one or more prestored threshold values represent predefined ranges of metabolic health parameters, wherein the predefined ranges comprise normal range or abnormal range for the metabolic health parameter.

16. The method (300) as claimed in claim 12, comprising:
receiving, through the wearable device (100), one or more user inputs for the metabolic health parameter at one or more predefined time intervals, wherein the one or more predefined time intervals comprise a fasting time interval to record metabolic health parameter, and one or more post prandial time intervals to record the metabolic health parameter, wherein the one or more user inputs are received over a predefined periodic time;
comparing, through the processing circuitry (104), the one or more user inputs for the metabolic health parameter with the sensor data using one or more AI algorithms to identify a correlation between the one or more user inputs and the sensor data; and
calibrating, through the processing circuitry (104), the wearable device (100) based on the correlation.

17. The method (300) as claimed in claim 12, wherein the one or more AI algorithms are used for training the adaptive prediction model, and wherein the one or more AI algorithm comprises semi-supervised learning for mapping the sensor data with the one or more user inputs, deep neural networks for initial training of the adaptive prediction model, incremental learning for periodic updates on the metabolic health pattern, and feature engineering techniques for processing the sensor data to generate the metabolic health pattern.