Development and validation of multicentre study on novel Artificial Intelligence-based Cardiovascular Risk Score
Shiv Kumar Jalepalli1, Prashant Gupta2, Andre L A J Dekker3
1Apollo Hospitals, Hyderabad, Telangana, India.
Insights
A new AI-based cardiovascular disease (CVD) risk score demonstrates superior prediction accuracy for cardiac events in the Indian population compared to traditional methods like FHRS and QRisk3.
Area of Science:
- Cardiology
- Artificial Intelligence
- Public Health
Background:
- Cardiovascular diseases (CVD) account for nearly 30% of deaths in India.
- Existing CVD risk scores show limited performance and reproducibility in the Indian population.
- There is a need for improved risk prediction models tailored to Indian demographics.
Purpose of the Study:
- To develop an Artificial Intelligence-based Risk Score (AICVD) for predicting 10-year CVD events.
- To compare the predictive performance of AICVD against the Framingham Heart Risk Score (FHRS) and QRisk3.
- To validate the AICVD model using independent cohorts from India and the Netherlands.
Main Methods:
- A deep learning hazards model was developed using a multilayered neural network on a large dataset of 31,599 Indian participants.
- 21 risk factors were selected through a multistep process involving Spearman correlation and propensity score matching.
- The model was validated on independent retrospective cohorts and compared with FHRS and QRisk3.
Main Results:
- The AICVD model achieved a high Area Under the Curve (AUC) of 0.853 in the primary cohort.
- Validation showed AUCs ranging from 0.84 to 0.92, with significantly better positive likelihood ratios and accuracy than FHRS and QRisk3.
- AICVD outperformed the Framingham Heart Risk Model in a Netherlands cohort (AUC 0.737 vs 0.707).
Conclusions:
- The novel AI-based CVD Risk Score (AICVD) demonstrates superior predictive performance for cardiac events in the Indian population.
- AICVD offers a more accurate and reliable tool for cardiovascular risk assessment compared to conventional scores.
- This AI-driven approach holds significant potential for improving CVD prevention strategies in India.
Objective:
Cardiovascular diseases (CVD) are one of the most prevalent diseases in India amounting for nearly 30% of total deaths. A dearth of research on CVD risk scores in Indian population, limited performance of conventional risk scores and inability to reproduce the initial accuracies in randomised clinical trials has led to this study on large-scale patient data. The objective is to develop an Artificial Intelligence-based Risk Score (AICVD) to predict CVD event (eg, acute myocardial infarction/acute coronary syndrome) in the next 10 years and compare the model with the Framingham Heart Risk Score (FHRS) and QRisk3.
Methods:
Our study included 31 599 participants aged 18-91 years from 2009 to 2018 in six Apollo Hospitals in India. A multistep risk factors selection process using Spearman correlation coefficient and propensity score matching yielded 21 risk factors. A deep learning hazards model was built on risk factors to predict event occurrence (classification) and time to event (hazards model) using multilayered neural network. Further, the model was validated with independent retrospective cohorts of participants from India and the Netherlands and compared with FHRS and QRisk3.
Results:
The deep learning hazards model had a good performance (area under the curve (AUC) 0.853). Validation and comparative results showed AUCs between 0.84 and 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%). In the Netherlands cohort, AICVD also outperformed the Framingham Heart Risk Model (AUC -0.737 vs 0.707).
Conclusions:
This study concludes that the novel AI-based CVD Risk Score has a higher predictive performance for cardiac events than conventional risk scores in Indian population.
Trial Registration Number:
CTRI/2019/07/020471.
Related Concept Videos
01:10Pre-Procedural Guidelines for Assessing Blood Pressure
01:23Assessment of blood pressure in brachial artery(two-step method)