Cardiovascular diseases are a leading cause of death across the globe. (Representational)
An Artificial Intelligence-powered tool to more accurately predict heart disease risk among Indians and recommend modifiable changes to prevent heart attacks has been launched by Apollo Hospitals in partnership with Microsoft.
Dr Prathap C Reddy, Chairman of Apollo Hospitals Group, said tools for physicians to understand the probability of their patients developing heart disease are mostly based on Western data sets and don't take regional risk variations into account.
"This impacts their accuracy when applied in an Indian context. The Apollo AI-powered Cardiovascular Disease Risk tool will change that and put the knowledge and the means to predict and prevent heart disease for Indians in the physician's hands,'' Dr Reddy said.
Not just Apollo's doctors, the AICVD risk score tool is being made available to all government and private doctors. "We are happy to share it with the government for its use on a large scale for the country to reduce the burden of cardiovascular disease in the future,'' said Sangita Reddy, joint managing director of Apollo Hospitals group.
Most risk score tools for heart disease take into account information such as blood pressure, diabetes level, dyslipidemia (good and bad fats), smoking history and stress. The tool developed by Apollo Hospitals has also factored in over 20 local parameters.
"For example, chewing tobacco is prevalent only in India and a few other countries such as Malaysia. That has been brought in," says Dr Keren Priyadarshini, who looks after Microsoft's healthcare initiatives.
Dr J Shiv Kumar, who has headed the effort to develop what he calls an "atmanirbhar" tool, said its algorithms are based on 10 years of anonymised data of about 4,00,000 individuals across the country collected by the Apollo Hospitals team.
"We used Microsoft Azure to turn that anonymised data into actionable insights that data scientists and clinicians could use to train machine learning models. The tool has been validated using data from the Maastricht Study, a long-term cohort health study in the Maastricht region and from the Heart+Vascular Center at Maastricht University Medical Center, Netherlands, '' Dr Kumar said.
"The accuracy in predicting disease over 10 years could be to the extent of nearly 90 per cent,'' he added.
There are three risk categories -- high, moderate and minimal. The tool also gives insights on risk contributors that can be modified to improve the score.
"If you are in the high-risk category, you could repeat the risk score assessment every 3-6 months to know if the changes you have made have helped you progress or you have regressed,'' Dr Kumar explained.
"You can get your risk assessment done by paying Rs 200 at Apollo, elsewhere too, the doctor's charges may be added, but we are sharing with them free of cost. They can access it by sending details to infoaicvd@apollohospitals.com,'' Ms Reddy said.
Dr Prathap Reddy thanked the Prime Minister for the recently-announced Digital Health Mission and said the nation's economy is intricately linked to the health of the people. Non-communicable diseases are estimated to account for an economic burden of 5% to 10% of GDP and cardiovascular disease tops the list, Dr Reddy said.
"Indians are genetically more predisposed and vulnerable to heart disease; India is the diabetes capital of the world. Though non-modifiable factors cannot be controlled, risk can be reduced by working on modifiable risk factors," said Dr Kumar.
Prof Rishi Sethi of Lucknow's King George Medical University said that a retrospective study on 1300 -1400 patients under 40, who suffered acute heart disease, showed that applying the AI-powered tool's risk score would have given much greater accuracy in predicting disease than any existing tool.
Cardiovascular diseases are the leading cause of death not just in the Asia Pacific, but across the globe, except in Africa, said Dr Azhar Ali, managing partner of Forth Medical Group, Scotland. He added that the advantage of the new tool was that it could be adapted for any population as per local risk factors.