lancet-header

Preprints with The Lancet is part of SSRN´s First Look, a place where journals identify content of interest prior to publication. Authors have opted in at submission to The Lancet family of journals to post their preprints on Preprints with The Lancet. The usual SSRN checks and a Lancet-specific check for appropriateness and transparency have been applied. Preprints available here are not Lancet publications or necessarily under review with a Lancet journal. These preprints are early stage research papers that have not been peer-reviewed. The findings should not be used for clinical or public health decision making and should not be presented to a lay audience without highlighting that they are preliminary and have not been peer-reviewed. For more information on this collaboration, see the comments published in The Lancet about the trial period, and our decision to make this a permanent offering, or visit The Lancet´s FAQ page, and for any feedback please contact preprints@lancet.com.

Multicenter Retrospective and Comparative Study on a Novel Artificial Intelligence Based Cardiovascular Risk Score (AICVD)

35 Pages Posted: 29 Mar 2020

See all articles by Shivkumar Jallepalli

Shivkumar Jallepalli

Apollo Hospitals

Prashant Gupta

Microsoft Corporation - Microsoft India Development Centre

Sujoy Kar

Apollo Hospitals

More...

Abstract

Introduction: Cardiovascular diseases (CVD) are one of the most prevalent diseases in India amounting for nearly 30% of total deaths. Dearth of research on CVD risk scores in Indian population, limited performance of conventional risk scores and the inability to reproduce the initial accuracies in randomized clinical trials – has led to the study on a large-scale patient data to determine and analyse risk scores with high accuracy and precision.

Objective: Objective of this study is to develop an Artificial Intelligence based Risk Score (AICVD) to predict CVD Event (e.g. Acute MI / ACS) in next 7 years and compare the model with risk scores like Framingham Heart Risk Score (FHRS) and QRisk3.

Methodology: Initial study included 31,599 participants aged 18 to 91 years from 2010 to 2017 from 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 Leant 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 3246 participants and compared with FHRS and QRisk3.

Results: The performance of the Deep Learnt Hazard model was at 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%).

Conclusion: The study concludes that the novel AI based CVD risk score improves on accuracy and precision of prediction than conventional risk scores. The use of deep learning reflects on interplay of multiple risk factors and provides an accurate and precise stratification of Cardiovascular Disease risk.

Trial Registration: Clinical Trial Registry of India (CTRI) - CTRI/2019/07/020471

Funding Statement: None.

Declaration of Interests: All authors declare that there are no competing / conflicting interests.

Ethics Approval Statement: The prospective research has been approved in all of nine center’s Institutional Ethics Committees and had been deliberated with over 100 Cardiologists nationally and internationally.

Keywords: Cardiac Risk Factors and Prevention; Cardiovascular Risk Score; Artificial Intelligence; Validation Study

Suggested Citation

Jallepalli, Shivkumar and Gupta, Prashant and Kar, Sujoy, Multicenter Retrospective and Comparative Study on a Novel Artificial Intelligence Based Cardiovascular Risk Score (AICVD) (March 4, 2020). Available at SSRN: https://ssrn.com/abstract=3548790 or http://dx.doi.org/10.2139/ssrn.3548790

Shivkumar Jallepalli

Apollo Hospitals ( email )

Jubilee Hills Hyderabad
India

Prashant Gupta

Microsoft Corporation - Microsoft India Development Centre ( email )

Hyderabad
India

Sujoy Kar (Contact Author)

Apollo Hospitals ( email )

Jubilee Hills Hyderabad
India