Description:4. DESCRIPTION:
Field of the invention:
The present disclosure generally relates to the technical field of cardiovascular disease prediction systems and in specific, relates to a system that measures the cardio vascular risk score of an individual to predict the risk of cardio vascular related diseases in next 10 years.
Background of the invention:
A cardiovascular disease (CVD) represents heart and blood vessel diseases that include ischemic heart diseases (problem with the circulation of blood to heart muscle), cerebrovascular diseases (problem with circulation of blood in the blood vessels of the brain) and peripheral vascular diseases (affecting the circulation primarily in the legs). The cardiovascular disease develops a number of complications including fatal or non-fatal myocardial infarction, stroke, angina pectoris, transient ischemic attacks and peripheral arteriopathy thereof.

Initially, at twentieth century, cardiovascular disease was responsible for 10% of all deaths worldwide. But, now-a-days it is responsible for about 30% of all deaths worldwide and 80% of these deaths occur in the developing countries. The cardiovascular disease, besides being the leading death cause, it is a highly prevalent disease that causes high health care costs. From the public health point of view, a policy was developed in relation to the cardiovascular disease that seeks to reduce the population’s risk of developing cardiovascular disease.

To attain the reduction in population risk, the stratification of the population in relation to the cardiovascular risk allows the establishment of preventive measures to prevent or delay the onset of the disease. The stratification helps in treating afflicted subjects by improving efficiency i.e., avoiding the occurrence of cardiovascular events and complications and cost-effectiveness. The existing risk factors such as dyslipidemia (that elevates LDL-cholesterol), hypertension, diabetes, consumption of tobacco and sedentary lifestyle are the direct causes of coronary diseases. These risk factors are common in universal population i.e., in almost every geographic region.

In existing technology, a system and method for bio-monitoring and personalized healthcare is known. At Apollo Hospitals’ network, the healthcare systems leverage technology to build integrated healthcare delivery models, which facilitate seamless electronic medical records. Artificial intelligence, machine learning and the deep-learning, in particular, is empowering the use of labelled clinical data from these electronic medical records – ‘big’ in terms of volume, variability, velocity, or scalability - with significantly enhanced computing power and cloud storage. At Apollo Hospitals’ clinical practice and population health perspective, this is making initial steps to have an impact at these fundamental levels:
For patients, by enabling them with better decisions for access and to process their own data, federated – secured – meaningfully used to promote health,
For clinicians, predominantly via supported, accurate interpretation of patient data including electronic health records and images;
For hospitals, by improving throughput, enhancing patient safety and the potential for reducing healthcare cost; and
For technology providers – constructing digital health platforms and creating positive network effects where patients, providers, payors and physicians can derive intense clinical value, and lastly
For Community – by creating solutions that are accessible and can reach the last mile.

In updated technology, a computer method is used to access an individual data for predicting the risk of cardiovascular event over 10 year period. This is done by - Differentiating Risk Factors in the Studied Population:
Concept of Assessing Holistic Risk through Machine Learning: Interplay (or association) of multiple risk factors are shown to have more significance in predicting an event than any individual factor. The hazard ratios associated with risk factors have demonstrated this in the current study. The study showed higher significance for risk factors like uncontrolled diabetes (Hazard Ratio – 2.32), hypertension (HR – 1.54) & dyslipidaemia (1.16) and chewing tobacco (2.01).
Differentiating approach of the Designed Risk Score:
Deep Survival Model : deep learning methods used in this study are different from traditional rule-based approaches (e.g. Framingham Risk Score) – which is novel and clinically differentiating.
Feedback loop: integrated with the institutional Electronic Medical Record (EMR) system. This would provide an ongoing prospective feedback loop for individuals over period of next few years, tracking their progress and their clinical data would continue to help machine learn and improve.

Hence, existing systems calculate present health status of the patient but could not predict future health conditions of the patient. The systems restrict to patients who suffer from coronary heart disease and predict the health status of the patient only for around five years. In addition, such systems do not provide the patient’s risk status and do not provide any predicted health status for an individual based on present conditions.

Therefore, there is a need for a system that can predict future health conditions of the patient and the individuals based on present health status. Apollo Hospitals developed and implement an AI CVD model using deep learning. The model concludes that the novel AI based CVD risk score has an improved predictive performance than conventional risk scores.The use of deep learning shows the interplay of multiple risk factors (predictors) and provides an accurate and precise stratification of Cardiovascular Disease risk.
Objectives of the invention:
The primary objective of the invention is to provide systems and methods for measuring a cardio vascular risk score of an individual to predict the risk of cardio vascular related diseases in next 10 years.

The other objective of the invention is to provide systems and methods for predicting the patient’s risk stage – Low, Moderate or High.

Another objective of the invention is to provide systems and methods forpredicting health condition of the patient with 90% accuracy.

The other objective of the invention is to provide systems and methods for generating a patient’s risk score report with intervention requirement based on the predicted risk category.

Yet another objective of the invention is to provide systems and methods for measuring a patient risk score based on present health status.

Further objective of the invention is to provide systems and methods for calculating an individual risk score based on previous patients record.
Summary of the invention:
The present disclosure proposes a system for predicting a cardiovascular disease using artificial intelligence (AI). The following presents a simplified summary in order to provide a basic understanding of some aspects of the claimed subject matter. This summary is not an extensive overview. It is not intended to identify key/critical elements or to delineate the scope of the claimed subject matter. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented later.

In order to overcome the above deficiencies of the prior art, the present disclosure is to solve the technical problem to provide a system that measures the cardiovascular risk score of an individual to predict the risk of cardiovascular related diseases in future.

According to an aspect, the invention provides a system for predicting the cardiovascular disease. The system for predicting the cardiovascular disease comprises a data acquisition module and a processing module. The system evaluates the CVD risk score and predicts the chances of having the cardiovascular related disease in the future.

The data acquisition module is configured to collect multiple health records of multiple users from a database. In specific, the database is stored with multiple health records of multiple users which are gathered from multiple hospitals at various areas. The processing module is configured to process the collected multiple health records to calculate a cardiovascular disease (CVD) risk score of at least one user.

The evaluated CVD risk score is categorized into at least one of a minimal risk, a moderate risk and a high risk thereof. The processing module comprises a correlation module and a prediction module. The correlation module is configured to correlate the collected multiple health records of multiple users with at least one medical record of multiple patients having discharge summaries of the CVD. In specific, the multiple health records comprise at least one of demographic details, personal data, physical data, heart health attributes, lifestyle and medical history thereof.

The prediction module is configured to select multiple clinical parameters that include Age, BMI, diastolic / systolic blood pressure, pulse rate, Alcohol, smoking and tobacco use were categorized into current, past or non – consumer, History of Diabetes mellitus, Hypertension, Dyslipidaemia, Gender, Diet (Vegetarian) and physical activity were used as features thereof from the correlated multiple health records and predict the CVD risk score of at least one user for the next decade using artificial intelligence (AI) algorithms that include a deep learning classification model thereof.

The evaluated CVD risk score is stratified and provided with individualized protocols using a clinical decision support system on the next best actions with an accuracy of above 90%. The individualized protocols comprise at least one lab, imaging and investigations, at least one cardiology referral, treatment goals, education and revisiting guidelines thereof.

According to another aspect, the invention provides a method for predicting the cardiovascular disease (CVD). The model included 31,599 participants aged 18-91 years from 2009 - 2018 in six Apollo Hospitals in India. A multi-step risk factors selection process using Spearman correlation coefficient and propensity score matching yielded 21 risk factors. A Deep Learning Hazard Model was built on risk factors to predict event occurrence (classification) and time to event (hazard model) using multi-layered neural network. Further, the model was validated with independent retrospective cohorts of participants from India and the Netherlands and compared with FHRS and QRisk3.

The Deep Learning Hazard model had a good performance (AUC 0.853). Validation and comparative results showed AUCs between 0.84 to 0.92 with better Positive Likelihood Ratio (AICVD-6.16 to FHRS–2.24 and QRisk3–1.16) and Accuracy (AICVD– 80.15% to FHRS 59.71% and QRisk3 51.57%).

Next, multiple clinical parameters are selected from the correlated multiple health records and the CVD risk score of at least one user is predicted for the next decade using AI algorithms. Finally, the CVD risk score is evaluated and the chances of having the cardiovascular related disease in the future are predicted. In specific, the evaluated CVD risk score is stratified and provided with individualized protocols using a clinical decision support system on the next best actions with an accuracy of above 90%.

Further, objects and advantages of the present invention will be apparent from a study of the following portion of the specification, the claims, and the attached drawings.
Detailed description of drawings:
The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate an embodiment of the invention, and, together with the description, explain the principles of the invention.

FIG. 1 illustrates an exemplary diagram of a system for predicting a cardiovascular disease (CVD) using artificial intelligence, in accordance to an exemplary embodiment of the invention.

FIG. 2 illustrates an exemplary method of a system for predicting the cardiovascular disease (CVD) using artificial intelligence, in accordance to an exemplary embodiment of the invention.

FIG. 3 illustrates an exemplary diagram of a system for development, validation and comparison process for CVD study, in accordance to an exemplary embodiment of the invention.

FIG. 4 illustrates an architecture of a deep learning model used in the development of CVD, in accordance to an exemplary embodiment of the invention.

FIG. 5 illustrates an architectural design clinical AI application programming interface (API), in accordance to an exemplary embodiment of the invention.

FIG. 6 illustrates a comprehensive data flow summary of CVD prediction, in accordance to an exemplary embodiment of the invention.

FIG. 7A and FIG. 7B illustrate the results of a cox proportional hazard model and a deep learning model, in accordance to an exemplary embodiment of the invention.

FIG. 7C and FIG. 7D illustrate the validation and comparison of two developed cities, in accordance to an exemplary embodiment of the invention.
Detailed invention disclosure:
Various embodiments of the present invention will be described in reference to the accompanying drawings. Wherever possible, same or similar reference numerals are used in the drawings and the description to refer to the same or like parts or steps.
The present disclosure has been made with a view towards solving the problem with the prior art described above, and it is an object of the present invention to provide a system that measures the cardiovascular risk score of an individual to predict the risk of cardiovascular related diseases in future.
According to an exemplary embodiment of the invention, FIG. 1 refers to a system 100 for predicting a cardiovascular disease (CVD). The system 100 for predicting the cardiovascular disease comprises a data acquisition module 102 and a processing module 104. The system 100 evaluates the CVD risk score and predicts the chances of having the cardiovascular related disease in future.
The data acquisition module 102 is configured to collect multiple health records of multiple users from a database. In specific, the database 112 is stored with multiple health records of multiple users which are gathered from multiple hospitals at various areas through a server. The processing module 104 is configured to process the collected multiple health records to calculate the cardiovascular disease (CVD) risk score of at least one user.
The evaluated CVD risk score is categorized into at least one of minimal risk, moderate risk and a high risk thereof based on given risk scores. In specific, the user CVD risk score is divided by the optimal score, if the relative score is less than 1x then the risk score is categorized as minimal risk. If the relative score is between 1x and 1.5x then the risk score is categorized as moderate risk. If the relative score is greater than 1.5x then the risk score is categorized as high risk.
The processing module 104 comprises a correlation module 106 and a prediction module 108. The correlation module 106 is configured to correlate the collected multiple health records of multiple users with at least one medical record of multiple patients having discharge summaries of the CVD. In specific, the multiple health records comprise at least one of demographic details, personal data, physical data, heart health attributes, lifestyle and medical history thereof.
The prediction module 108 is configured to select multiple clinical parameters that include age, gender, BMI, systolic blood pressure and smoking habits thereof from the correlated multiple health records and predict the CVD risk score of at least one user for the next decade using artificial intelligence (AI) algorithms that include at least one of a machine learning model, a cox proportional hazard model and a deep learning classification model thereof. In specific, the evaluated CVD risk score is stratified and provided with individualized protocols using a clinical decision support system on the next best actions with an accuracy of above 90%. The individualized protocols comprise at least one lab, imaging and investigations, at least one cardiology referral, treatment goals, education and revisiting guidelines thereof.
According to another exemplary embodiment of the invention, FIG. 2 refers to a method for predicting the cardiovascular disease (CVD). At step 202, the multiple health records of multiple users are collected from a database by using the data acquisition module. At step 204, the multiple health records of the multiple users are correlated with at least one medical record of multiple patients having discharge summaries of the CVD.
At step 206, multiple clinical parameters are selected from the correlated multiple health records and a CVD risk score of at least one user is predicted for a next decade using AI algorithms. At step 208, the CVD risk score is evaluated and the chances of having the cardiovascular related disease in the future are predicted. In specific, the evaluated CVD risk score is stratified and provided with individualized protocols using the clinical decision support system on next best actions with an accuracy of above 90%.
According to another exemplary embodiment of the invention, FIG.3 refers to diagram of the development, validation and comparison process 300 for the CVD study. At step 302, around 40,000 individuals’ preventive health check (electronic medical) records are collected between 2009 to 2018 from various cities of multiple Hospitals. At step 304, the collected data is matched with adult patient discharge summaries with CVDs (ACS/AMI). However, cardiac admissions, heart failure, and rheumatoid disease are excluded.
At step 306, data hygiene for multiple clinical features for a health check and patient’s discharge summaries are selected. Later, employing CCA, inappropriate data, blank cells or ambiguity is excluded. At step 308, patients with a diagnosis of CVD and at least one health check record in past 10 years are selected. Simultaneously, individuals with at least two health check records in the past 10 years and no documented CVD are selected. Later, individuals with two health check and documented evidence of no CVD is excluded.
At step 310, feature optimization and selection is carried out using correlation coefficients. Later, non-contributing clinical features are removed. In specific, the selected features range comprise from >=0.3 and =<-0.3. At step 312, deep neural network (four layers), cox progression hazard model and classification layer are added. In specific, AUC – 0.853 + AUC – 0.83. At step 314, validate selected risk factors clinically, checking the p-value and hazard ratio. At step 316, the retrospective validation is done between the hospital’s data and medical institutions data. At step 318, comparison with high-risk scores is carried out based on sensitivity, heart risk score and thereof.
According to another exemplary embodiment of the invention, FIG. 4 refers to a diagram 400 of risk interplay with deep learned hazard model. Input to the model is the normalized predictors (between 0 & 1), where each neuron represents each risk factor. This is followed by 4 nodes of deep neural network of size 100, 70 40 & 35 respectively. 50 Epochs – 4 layers are used with corresponding AUC, patient review and coordination (PRC), accuracy and loss which are noted at each epoch step and found to provide better output than other combinations (like 3 or 5 layers with different Epochs combination).
At the final layer, the major challenge to merge classification & regression loss can be achieved. As classification loss is numerical value (difference between the number of days between actual & predicted days) while binary classification loss would be between 0 & 1 (No Event & Event). Further, computing of both regression & classification loss can be carried out using the same ranges and added Max scaling based loss, where regression loss is scaled down between 0 and 1. The final algorithm based on this calculation is used to build the deep survival model and an application programming interface to calculate the risk score.
Input to the model is normalized risk factors (between 0 and 1), where each neuron represents each risk factor. This is followed by 4 layers of deep neural network of sizes 100, 70, 40 and 25 respectively.
RL1 = FC100(Input)∈ℝ100 (1)
RL2=FC70(RL1)∈ℝ70 (2)
RL3=FC40(RL2)∈ℝ40 (3)
RL4=FC25(RL3)∈ℝ25 (4)
As the next step, a regression layer is added, where this layer is found from survival days (number of days before cardiac event).
CoxLayer=FC1(RL4)∈ℝ1 (5)
Multiple approaches for regression loss calculation, including Root Mean Square Error loss as well as Maximum Likelihood loss has been carried out, but the better accuracy numbers are achieved by using negative log partial likelihood and derivation. The equations for the same suggested by them are as follow (Loss 1):

Gradients for back propagations are calculated using the following equation:

Where X_i are inputs to the output layer from RL4 (equation 4), β are co-efficients for Cox parameters (described as fully connected layer in equation 5), U is set of uncensored samples, R_i is the set of “at-risk samples” where follow-up time Y_j>Y_i. One of the issue with current deep survival networks are to not utilize censored data in the model. To utilize right censored data properly, we added a classification layer parallel to Cox-Regression layer and calculated Binary loss via binary cross entropy loss (Loss 2).
Classification Layer = FC2(RL4)∈ℝ (8)
Cross Entropy Loss = −(y log(p) + (1 − y) log(1 − p)) (9)
One of the major challenges is to merge classification and regression loss. As regression loss would be numerical value (difference between the number of days between actual and predicted days) while binary classification loss would be between 0 and 1. One of the ways to handle this by adding a static weight score to both the loss, but it is observed that regression loss decreases drastically compared to classification loss after the first few iterations, so assigning pre-decided weights to both the loss terms are not helpful.
Changing the weights dynamically after every iteration would mean changing the error space and would mean that the model may not converge. To solve this problem, different approaches are tried. But models are based on Bayesian deep learning methodology and require learning two additional noise parameters for regression and classification loss.
These additional parameters increase/decrease inversely in proportion to their respective regression and classification loss. Based on multiple experiments we observed that similar objects can be achieved if both regression and classification loss are in the same ranges. Hence, finally, we added max scaling-based loss, where regression loss is scaled down between 0 and 1 and the average of Regression and Classification loss is used for further processing.
According to another exemplary embodiment of the invention, FIG. 5 refers to exemplary architectural design clinical AI application programming interface (API) 500. The multiple user data is collected from multiple sources that include multiple applications incorporated in various devices. The user data is stored in metadata management and thereby ET/ELT and transformation and harmonization.
The stored data is sent to machine learning (ML) modelling that includes statistical tools and methods and machine learning tools thereof. The data is validated using application programming interface (API) and output validation that comprises application programming interface (API) process, clinical algorithm, API-based validation or verification and federated learning thereof. The validated data is used by digital health platforms, cloud marketplace, standard web mobile applications and hospitals thereof.
According to another exemplary embodiment of the invention, FIG. 6 refers to the exemplary diagram 600 of the comprehensive data flow summary. The data is collected manually captured and stored in central servers from various hospitals of multiple cities. In specific, data is captured electronically and stored in central servers. The data includes two or more health check visits and CVD discharges. The stored data is extracted by the analytics team. In specific, the analytics team correlates coefficients with clinical parameters using a deep neural network and thereby predicts a CVD risk score with user personal data that includes name, age and BMI thereof.
According to another exemplary embodiment of the invention, FIG. 7A and FIG. 7B refer to the results of a cox proportional hazard model and a deep learning model. Initially, multiple predictors are selected through the process of correlation coefficient methods. These reflect the details of various parameters and their incidence in different age and gender groups. Later, propensity score matching was performed for all variable on the model set. The values between 0.2 - 0.4 are taken as very good fit of the model.
Overall 22.19% cases had CVD events. The occurrence of the CVD events was 40.01 per 1000 persons. In specific, the CVD events occurrence was 41.43 per 1000 person’s years in male and 36.78 per 1000 person’s years in female. The cox proportional hazard approach has an accuracy of approximately 0.83, whereas the deep survival method has an accuracy of around 0.853. The precision-recall curve indicates that the deep learning approach is very accurate in identifying those at high risk, with AP = 0.73 in the deep learning model vs. AP = 0.67 in the cox proportional hazard model.
The calibration slope depicts that the hazard model may misses few high risk patients. As a result, the combination of both models is the most successful method. The area under the curve (AUC) of the deep survival model is three percentage points higher than that of the cox-proportional model. The extra predictors earlier increased the AUC for the deep survival model by 1.5%. The factors are listed in order of hazard ratio and associated p value with 95% - upper and lower confidence intervals (CI).
Diabetes mellitus (HR: 2.342 (2.32 – 2.36); p value <0.001), hypertension (HR: 1.543 (1.52 – 1.57); p value <0.001), diastolic blood pressure (1.065 (1.04 – 1.09); p value <0.001), chewing tobacco (HR: 2.01 (1.99 – 2.036); p value <0.001), smoking (HR: 2.277 (2.26 – 2.299); p value <0.0001) and Dyslipidaemia (1.16 (1.14 – 1.18); p value <0.001) emerge as the most significant cardiovascular risk parameters.
Multiple parameters improve the cardiovascular risk events, including age variation, which accounts for about 4% of the risk factor, BMI variation, which accounts for 7% of the risk factor, uncontrolled diabetes, which has a 2.34 times higher risk, and hypertension, which has a 50% risk of CVD events. Raised diastolic blood pressure (6.57%) is associated with a somewhat greater risk than elevated systolic blood pressure (2.5%). Individuals who smoke or chew tobacco are twice as likely, while those who have dyslipidaemia are 16% more likely.
According to another embodiment of the invention, FIG. 7C and FIG. 7D refer to validation and comparison of two developed cities. The AUC scores of validation corresponds of the initial model are 0.844 for Delhi and 0.921 for Kolkata. In specific, establishing the high precision (0.76 and 0.87), recall (0.80 and 0.84) and F1 score (0.77 and 0.85) respectively for Delhi and Kolkata. The negative span for Delhi hospitals was >=1800 days and for Kolkata hospitals was >=1000 days. This indicates that the frequency is inversely proportional to accuracy. For example, the lower the frequency, the higher the accuracy.
The Sensitivity and Specificity of the AICVD (61.31% & 90.04%) is higher than FHRS (37.63% & 82.70%) and QRisk3 (31.03% & 76.08%) respectively. Positive Likelihood Ratio in AICVD score is at 6.16 (5.37 to 7.06) signifying moderate to large accuracy in probability of predicting cardiovascular Event compared to FHRS 2.18 (1.92 to 2.46) and QRisk3 1.30 (1.16 to 1.45) in validation set. Similarly, Negative Likelihood Ratio is at 0.43 in AICVD compared to Framingham (0.75) and QRisk3 (0.91).
The application of the AICVD Risk Score (Mid and High Risk) in this dataset to the Maastricht population resulted in a precision of 0.94 and a recall of 0.62. The model achieved 73.7% area under the ROC curve, compared to 70.7% using the Framingham Risk Score. The slope of the calibration plot was somewhat less than one, implying that the chance of a cardiovascular event was slightly overestimated among individuals at greater risk.
The Algorithm's Application Programming Interface (API) takes the data specified in the Methodology section and generates four specific outputs: an individual's risk score and optimal score for age and gender, the top three modifiable risk factors, a trend line of risk scores over time and a Clinical Algorithm that assists the Physician in taking next steps based on an individual's risk score.
Numerous advantages of the present disclosure may be apparent from the discussion above. In accordance with the present disclosure, a system for predicting cardiovascular disease using artificial intelligence is disclosed here. The proposed system and method measure a cardiovascular risk score of an individual to predict the risk of cardio vascular related diseases in future. The proposed system and method predict the patient’s risk stage.
The proposed system and method predict the health condition of the patient with 90% accuracy. The proposed system generates patient’s risk score report with medication requirements based on the predicted risk category. The proposed system and method measure a patient risk score based on present health status. The proposed system and method calculates an individual risk score based on previous patients' records.
It will readily be apparent that numerous modifications and alterations can be made to the processes described in the foregoing examples without departing from the principles underlying the invention, and all such modifications and alterations are intended to be embraced by this application.
, Claims:
5. CLAIMS:
I / We Claim:
1. A system for predicting a cardiovascular disease, comprising:
a data acquisition module configured to collect user data of at least one user from multiple users in a database;
a processing module configured to process said collected data to calculate a risk score of at least one user; wherein said processing module comprises:
a correlation module configured to map said collected data of said at least one user with at least one medical record of multiple patients having discharged summaries of cardiovascular disease;
a prediction module configured to select multiple clinical parameters from said correlated health checked record and predict risk score of said at least one user for next decade using AI algorithms,
whereby said system evaluates said risk score and predicts said cardiovascular related disease in future.
2. The system as claimed in claim 1, wherein said artificial intelligence (AI) algorithms comprises at least one of machine learning model, a cox proportional hazard model and a deep learning classification model thereof.
3. The system as claimed in claim 1, wherein said evaluated risk score is stratified and individualised protocol is provided using a clinical decision support system or next best action with an accuracy of 90%.
4. The system as claimed in claim 3, wherein said individualised protocol provide at least one recommended protocol that comprises lab, imaging and investigation, at least one cardiology referral, treatment goals, education and revisit guidelines thereof.
5. The system as claimed in claim 1, wherein said evaluated risk score is categorized into at least one of relative risk that includes minimal risk, moderate risk and high risk thereof.
6. The system as claimed in claim 1, wherein said database comprises preventive health check data of plurality of individuals that are gathered from various area of multiple hospitals.
7. The system as claimed in claim 1, wherein said multiple clinical parameters include age, gender, BMI, systolic blood pressure and smoking habits thereof.
8. The system as claimed in claim 1, wherein said health check record comprises at least one of demographic details, personal data, physical data, heart health attributes, life style and medical history thereof of said user obtained from patient consultant.
9. A method for predicting a cardiovascular disease, comprising:
collecting user data of at least one user from multiple users in a database through a data acquisition module;
mapping said collected data with at least one medical record of multiple patients having discharged summaries of cardiovascular disease through a correlation module;
selecting multiple clinical parameters from said correlated health checked record and predicting a risk score of said at least one user for next decade using AI algorithms through a prediction module, and
evaluating said risk score and predicting said cardiovascular related disease in future through a processing module.
10. The method as claimed in claim 1, wherein said evaluated risk score is stratified and individualised protocol is provided using a clinical decision support system or next best action with an accuracy of 90%.
6. DATE AND SIGNATURE:
Dated this 10th day of September, 2022