Description:FIELD OF THE INVENTION
[0001] The invention is connected to the biometric systems of identification which utilize physiological signs to authenticate the user. To be more particular, it is a medical-grade biometric system that functions based on electrocardiogram (ECG) detection, fractal features extraction, and deep learning algorithms to recognize individuals in a healthcare and wearable context with accuracy and high security.
BACKGROUND FOR THE INVENTION:
[0002] The following discussion of the background to the invention is intended to facilitate an understanding of the present invention. However, it should be appreciated that the discussion is not an acknowledgment or admission that any of the material referred to was published, known, or part of the common general knowledge in any jurisdiction as of the priority date of the application. The details provided herein the background if belongs to any publication is taken only as a reference for describing the problems, in general terminologies or principles or both of science and technology in the associated prior art.
[0003] Common interactive biometric systems include fingerprint, facial recognition and iris scanning which are traditionally used to identify with a certain degree of security. There are however a number of challenges with these methods: They may be swindled with photos, molds or imitations.
- They are sensitive to the environment (lighting, temperature, or surface condition).
- They can evolve with time either through aging, injuries or make-up.
[0004] Conversely, the biometrics based on ECG applies electrical signals created by the human heart which are unique and cannot be duplicated or counterfeited. The ECG signals are dynamic, meaning that it is not a false input of a living person. The current ECG biometric techniques primarily rely on time domain or frequency domain features. These characteristics are not necessarily capturing the complicated and self-similarity of ECG signals. The invention addresses these limitations through the application of fractal mathematics to the complexity of the ECG waveforms to identify the identity and integrate it with deep learning to provide robust and accurate identities.
[0005] In light of the foregoing, there is a need for the Medical biometric identification system using fractal and deep learning analysis of ECG signals that overcomes problems prevalent in the prior art.
OBJECTS OF THE INVENTION:
[0006] Some of the objects of the present disclosure, which at least one embodiment herein satisfies, are as follows.
[0007] The principal object of the present invention is to overcome the disadvantages of the prior art by providing the Medical biometric identification system using fractal and deep learning analysis of ECG signals.
[0008] An object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals that offers a framework and procedure of biometric identification with ECG signals, integrating the fractal feature extraction algorithm and deep learning analysis.
[0009] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the invention uses fractal dimension analysis (Higuchi, Katz, and Petrosian) to measure the self-similarity and complexity of ECG signals to give a more discriminating and invariant signature of each individual.
[0010] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the fractal mathematical modeling methodology and combines deep learning (CNN/LSTM/hybrid) models, allowing the model to learn both morphological and time-based patterns of ECG signals at the same time.
[0011] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the authentication system is far more resilient than traditional methods which were amplitude - or frequency-based.
[0012] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the system enables real-time ECG capture, and identity check which enables continuous or on-demand authentication, in the wearable and telemedicine applications.
[0013] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein since ECG is an active physiological channel, which displays the actual cardiac activity, the technique excludes the chances of spoofing (as is true of other biometric methods, like fingerprint or facial recognition) and provides genuine physiological authentication.
[0014] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the invention permits optional inclusion of time-domain (RR intervals) and frequency-domain features in addition to fractal features to increase classification accuracy and flexibility to a wide variety of datasets.
[0015] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the deep learning module is constructed to be modular - it can use CNN to analyze spatial patterns, LSTM to analyze temporal patterns, and hybrid networks to analyze complex multi-scale ECG features based on fractal patterns.
[0016] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the fractal representation space means that neural models do not need input dimensionality as much, allowing them to be used with low-power devices such as wearables and portable ECG monitors.
[0017] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals, wherein the essence of the invention is the combination of a mathematical idea (fractal geometry) and artificial intelligence (deep learning) to identify biomedical signals in an ECG - something not done before in Ecg biometrics.
[0018] Another object of the present invention is to provide the Medical biometric identification system using fractal and deep learning analysis of ECG signals that allows a safe entry into the hospital, wearable identification, and telemedicine verification, where traditional biometrics is not feasible.
[0019] Other objects and advantages of the present disclosure will be more apparent from the following description, which is not intended to limit the scope of the present disclosure.
SUMMARY OF THE INVENTION:
[0020] The present invention provides Medical biometric identification system using fractal and deep learning analysis of ECG signals.
[0021] The invention is regarding the new method of the identification of a person based on the electrical activity of his or her heart. The heart of every person beats a little differently, just like how each person has a unique fingerprint. Such minute variations in heartbeats are possible to record using a signal known as an ECG (electrocardiogram). The ECG captures the rise and fall of the electrical signal of the heart as it beats. This natural signal emitted by the heart is the one that is used to identify and confirm the identity of a person through the invention. The ECG is also a direct result of the body, which makes it impossible to duplicate, counterfeit, or contaminated by injuries or light, unlike fingerprints or facial recognition that can be either. This renders cheating or imitating quite impossible.
[0022] The system initially picks up the ECG signal of the individual with the aid of a small wearable device or even by means of simple sensors to the skin. Then the signal that has been recorded is cleaned to eliminate the undesired noise or interference. After the signal of the heart is clear, it is analyzed to identify patterns that render it distinct to an individual. The system tries to understand these patterns through a special form of mathematical observation known as the fractal analysis. Put simply, the fractal analysis assists in gauging the complexity or detail of something even though it appears irregular. To take an example, a tree branch appears to be rough and uneven, but the general pattern of the branch can be repeated in smaller bites - the self-repeating nature of this analysis is measured by the fractal analysis.
[0023] Likewise, the ECG of any individual possesses some self-recurring form in its wave patterns which is individual. Once these distinctive characteristics have been identified, artificial intelligence, referred to as smart computer learning, is used to identify the identity of each individual. The AI gets to learn examples of ECG signals of various individuals and learns how to distinguish between them. A trained system can recognize a person in a short period of time, all that it needs to do is to scan the heart signal. Take an example of a hospital where such a system is deployed at the entrance to the hospital, a doctor would be identified automatically on pressing a sensor pad and no ID cards or passwords are needed. Equally, a wearable device by a patient can constantly check the presence of the right individual using the wearable device, thus, guarding medical information and ensuring that medical data is not lost and that the information will be owned by the right individual.
[0024] The peculiarity of this invention is that it is based on two brilliant concepts one is the complex heart signal (fractal pattern) and the other is machine learning and intelligence to create a quick, safe, and reliable way of identifying humans. It is operable whether the lighting varies, faces are covered or even the hands wet, since it is only a matter of the natural beat of the heart.
[0025] Fractal-Based Feature Extraction for ECG Biometrics: Contrary to old ECG biometric systems that follow time or frequency domain, the invention uses fractal dimension analysis (Higuchi, Katz, and Petrosian) to measure the self-similarity and complexity of ECG signals to give a more discriminating and invariant signature of each individual.
[0026] Hybrid Fractal–Deep Learning Architecture: The proposed model is a novel system based on the fractal mathematical modeling methodology and combines deep learning (CNN/LSTM/hybrid) models, allowing the model to learn both morphological and time-based patterns of ECG signals at the same time.
[0027] Enhanced Robustness Against Noise and Variability: Fractal properties are also resilient to signal distortions, baseline wander and session-to-session variation and as such, the authentication system is far more resilient than traditional methods which were amplitude- or frequency-based.
[0028] Real-Time Continuous Authentication Capability: The system enables real-time ECG capture, and identity check which enables continuous or on-demand authentication, in the wearable and telemedicine applications.
[0029] Anti-Spoof and Live Biometric Nature: Since ECG is an active physiological channel, which displays the actual cardiac activity, the technique excludes the chances of spoofing (as is true of other biometric methods, like fingerprint or facial recognition) and provides genuine physiological authentication.
[0030] Multi-Domain Feature Integration: The invention permits optional inclusion of time-domain (RR intervals) and frequency-domain features in addition to fractal features to increase classification accuracy and flexibility to a wide variety of datasets.
[0031] Generalizable Deep Neural Framework: The deep learning module is constructed to be modular - it can use CNN to analyze spatial patterns, LSTM to analyze temporal patterns, and hybrid networks to analyze complex multi-scale ECG features based on fractal patterns.
[0032] Low Computational Overhead for Embedded Implementation: The fractal representation space means that neural models do not need input dimensionality as much, allowing them to be used with low-power devices such as wearables and portable ECG monitors.
[0033] Novel Fusion of Mathematical and AI Techniques: The essence of the invention is the combination of a mathematical idea (fractal geometry) and artificial intelligence (deep learning) to identify biomedical signals in an ECG - something not done before in Ecg biometrics.
[0034] Application-Driven Innovation: Particularly tailored to the medical setting, it allows a safe entry into the hospital, wearable identification, and telemedicine verification, where traditional biometrics is not feasible.
BRIEF DESCRIPTION OF DRAWINGS:
[0035] Reference will be made to embodiments of the invention, examples of which may be illustrated in accompanying figures. These figures are intended to be illustrative, not limiting. Although the invention is generally described in the context of these embodiments, it should be understood that it is not intended to limit the scope of the invention to these particular embodiments.
[0036] Figure1A and 1B: Medical Biometric Identification System;
DETAILED DESCRIPTION OF DRAWINGS:
[0037] While the present invention is described herein by way of example using embodiments and illustrative drawings, those skilled in the art will recognize that the invention is not limited to the embodiments of drawing or drawings described and are not intended to represent the scale of the various components. Further, some components that may form a part of the invention may not be illustrated in certain figures, for ease of illustration, and such omissions do not limit the embodiments outlined in any way. It should be understood that the drawings and the detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the invention is to cover all modifications, equivalents, and alternatives falling within the scope of the present invention as defined by the appended claim.
[0038] As used throughout this description, the word "may" is used in a permissive sense (i.e. meaning having the potential to), rather than the mandatory sense, (i.e. meaning must). Further, the words "a" or "an" mean "at least one” and the word “plurality” means “one or more” unless otherwise mentioned. Furthermore, the terminology and phraseology used herein are solely used for descriptive purposes and should not be construed as limiting in scope. Language such as "including," "comprising," "having," "containing," or "involving," and variations thereof, is intended to be broad and encompass the subject matter listed thereafter, equivalents, and additional subject matter not recited, and is not intended to exclude other additives, components, integers, or steps. Likewise, the term "comprising" is considered synonymous with the terms "including" or "containing" for applicable legal purposes. Any discussion of documents, acts, materials, devices, articles, and the like are included in the specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention.
[0039] In this disclosure, whenever a composition or an element or a group of elements is preceded with the transitional phrase “comprising”, it is understood that we also contemplate the same composition, element, or group of elements with transitional phrases “consisting of”, “consisting”, “selected from the group of consisting of, “including”, or “is” preceding the recitation of the composition, element or group of elements and vice versa.
[0040] The present invention is described hereinafter by various embodiments with reference to the accompanying drawing, wherein reference numerals used in the accompanying drawing correspond to the like elements throughout the description. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiment set forth herein. Rather, the embodiment is provided so that this disclosure will be thorough and complete and will fully convey the scope of the invention to those skilled in the art. In the following detailed description, numeric values and ranges are provided for various aspects of the implementations described. These values and ranges are to be treated as examples only and are not intended to limit the scope of the claims. In addition, several materials are identified as suitable for various facets of the implementations. These materials are to be treated as exemplary and are not intended to limit the scope of the invention.
[0041] The present invention provides Medical biometric identification system using fractal and deep learning analysis of ECG signals.
[0042] The Medical Biometric Identification System Using Fractal and Deep Learning Analysis of ECG Signals is an integrated biometric authentication architecture developed to identify individuals based on the inherently unique electrical activity of the human heart. The system captures ECG signals through medical-grade or wearable sensors and preprocesses them to create a noise-free cardiac waveform representation. It then computes non-linear fractal features that quantify the personal complexity of a user’s heartbeat and further classifies the identity using a deep learning architecture optimized for spatiotemporal ECG interpretation. The final authentication module verifies the user’s identity in real time, making the system suitable for hospital access control, secure telemedicine, continuous wearable authentication, and high-security medical or industrial applications. The complete system functions as a combination of hardware modules for sensing and communication, embedded firmware for preprocessing, cloud or local deep learning engines for classification, and user-facing software for operational deployment.
[0043] The physical architecture includes an ECG acquisition unit equipped with clinical electrodes, wrist-based conductive pads, or disposable ECG patches capable of capturing signals at 250–500 Hz to extract precise P-QRS-T wave morphology. These sensors work alongside an analog front-end circuit comprising instrumentation amplifiers, band-pass filters, and analog-to-digital converters. An embedded microcontroller, typically an ARM Cortex-M series or equivalent, digitizes and packages the signals, after which a communication interface such as Bluetooth Low Energy, Wi-Fi, or USB transmits ECG packets to the processing hardware. A preprocessing processor unit, which may be built into a wearable device or implemented on a mobile, desktop, or cloud module, removes noise and artifacts using digital FIR or IIR filters, wavelet transformations, and adaptive threshold methods. It also detects QRS complexes using algorithms like Pan–Tompkins, Hilbert envelope analysis, or deep-QRS networks. The computation and classification unit may exist on-device or remotely and contains DSPs, CPUs, GPUs, and optional neural network accelerators for high-speed processing, while secure storage modules maintain encrypted biometric templates using AES-256 and medically compliant standards. The authentication interface then displays verification results and interacts with access control devices, medical software, wearable dashboards, mobile health applications, and patient monitoring systems.
[0044] The software architecture begins with a preprocessing module that transforms raw ECG signals into normalized and segmented cardiac cycles. It performs band-pass filtering across 0.5–40 Hz, removes baseline drift through wavelet decomposition or moving average methods, and suppresses noise using adaptive thresholding. R-peaks are detected for accurate beat segmentation, and amplitudes are normalized to maintain cross-device compatibility. Cleaned and segmented signals are then passed to the fractal feature extraction module, which forms the core novelty of the invention. This module computes Higuchi, Katz, and Petrosian fractal dimensions to capture irregularity, self-similarity, roughness, structural deviation, and micro-variation in the heartbeat. Optional features such as RR intervals, dominant frequency components, or wavelet energy coefficients may also be included. The resulting multi-dimensional feature vector serves as a unique and compact representation of an individual’s cardiac identity.
[0045] These fractal features are then fed to the deep learning module, which performs identity classification using a hybrid CNN–LSTM architecture. Convolutional layers extract morphological patterns from heartbeat windows or fractal vectors, while LSTM layers capture temporal characteristics across consecutive cardiac cycles. The hybrid architecture effectively learns multi-scale spatial and temporal identity cues, and its Softmax output layer assigns probability values to registered users. The model undergoes supervised training during enrollment and supports continuous refinement during real-world usage. The authentication module compares live features with stored templates, computes match probabilities, applies configurable thresholds, and outputs authentication results or prompts for signal reacquisition. The system can also carry out continuous authentication for wearable applications by reevaluating identity every few seconds.
[0046] The operational method begins when the user touches a sensor pad or wears an ECG device, enabling continuous ECG acquisition. The preprocessing stage removes noise, extracts stable cycles, and prepares the waveform for biometric computation. Fractal analysis then generates Higuchi, Katz, and Petrosian dimensions to form a distinctive fractal signature, which the deep learning model processes to assign identity probabilities. When the probability exceeds a secure threshold, the system confirms identity and grants access to hospital applications, wearable functions, or remote healthcare systems. Optional continuous monitoring maintains authentication persistently.
[0047] From the user’s perspective, the system operates seamlessly. A doctor approaching a hospital entrance need only touch a sensor pad for the system to identify them instantly without ID cards or passwords. A patient wearing a smart ECG patch is automatically authenticated each time health data is transmitted to a telemedicine server. Smartwatch users simply rest a finger on the electrode to unlock secure apps or medical records. Because ECG signals originate internally and cannot easily be spoofed or replicated, users are shielded from lighting issues, fingerprint contamination, or presentation attacks common in other biometric systems.
[0048] The invention offers wide industrial applicability. In healthcare, it enables secure hospital access, ICU identity verification, protected electronic health record retrieval, telemedicine authentication, and reliable identity assurance in remote monitoring systems. In wearables, it integrates seamlessly into smartwatches, fitness bands, implantable devices, and home diagnostic equipment. Security and defense sectors can use ECG authentication for high-security labs, communication devices, and personnel verification. Banking and fintech can embed ECG-secured transactions or mobile wallet access. Clinical research benefits from participant authentication across multicenter trials, preventing identity mismatches. IoT and smart infrastructures can also adopt physiological authentication for smart homes and medical IoT ecosystems.
[0049] The invention provides significant advantages, including intrinsic spoof resistance due to ECG’s internal generation, high noise robustness enabled by fractal features, and the ability to capture nonlinear heartbeat complexity not available in conventional time- or frequency-domain methods. The hybrid CNN-LSTM architecture achieves high accuracy by learning personalized ECG morphology, while compact fractal vectors keep computational load low, allowing real-time processing on embedded devices. Privacy is strengthened due to the difficulty of externally acquiring or stealing ECG data. The system is highly suitable for medical environments where identity verification and health monitoring coexist and is scalable across cloud, hospital, and wearable platforms.
[0050] Because of its modular design, the system can be integrated into hospital management systems, wearable firmware, government health infrastructures, military authentication equipment, mobile health applications, and IoT-based home healthcare devices. It supports multiple wired and wireless communication protocols including Bluetooth, Wi-Fi, GSM, 4G/5G, ZigBee, and hospital networks, enabling flexible integration across diverse technological ecosystems.
[0051] The disclosure has been described with reference to the accompanying embodiments herein and the various features and advantageous details thereof are explained with reference to the non-limiting embodiments in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein.
[0052] The foregoing description of the specific embodiments so fully revealed the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and/or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the scope of the embodiments as described herein. , Claims:We Claim:
1) A biometric identification system for authenticating an individual using electrocardiogram (ECG) signals, the system comprising:
- an ECG acquisition unit configured to capture cardiac electrical activity through one or more electrodes at a sampling rate between 250–500 Hz;
- a preprocessing module configured to remove noise, extract R-peaks, and segment heartbeats;
- a fractal feature extraction module adapted to compute non-linear fractal metrics including Higuchi Fractal Dimension, Katz Fractal Dimension, and Petrosian Fractal Dimension;
- a deep learning classification module comprising a hybrid CNN-LSTM architecture trained to classify the extracted features as belonging to a registered user; and
- an authentication module configured to compare live features with stored templates and generate an authentication result;
- wherein the system identifies the individual based on the unique fractal and morphological characteristics of the ECG signal.
2) The biometric identification system as claimed in Claim 1, wherein the fractal feature extraction module generates a multi-dimensional fractal signature vector combining fractal dimensions with optional physiological metrics including RR intervals, dominant frequency components, or wavelet energy coefficients.
3) The biometric identification system as claimed in Claim 1, wherein the preprocessing module performs band-pass filtering between 0.5–40 Hz, removes baseline drift using wavelet decomposition, applies adaptive noise suppression, and normalizes amplitude to achieve cross-device consistency.
4) The biometric identification system as claimed in Claim 1, wherein the deep learning classification module comprises one or more convolutional layers configured to extract spatial morphological features of the heartbeat and one or more LSTM layers configured to learn temporal variations across consecutive cardiac cycles.
5) The biometric identification system as claimed in Claim 1, wherein the system supports continuous authentication by re-evaluating the user identity at predetermined intervals during wearable or portable device usage.
6) The biometric identification system as claimed in Claim 1, wherein the secure storage module stores fractal feature templates in encrypted form using AES-256 or medically compliant data security standards.
7) The biometric identification system as claimed in Claim 1, wherein the ECG acquisition unit comprises medical-grade electrodes, wrist-based conductive pads, adhesive ECG patches, or integrated wearable sensors embedded in smartwatches or fitness devices.
8) The biometric identification system as claimed in Claim 1, wherein the authentication module outputs an identity status selected from: authenticated, rejected, or signal reacquisition required.
9) The biometric identification system as claimed in Claim 1, wherein the system is deployable as a hospital access control terminal, telemedicine authentication module, wearable device identity layer, secure EHR access system, or IoT-based physiological authentication interface.