Survey and Evaluation of Hypertension Machine Learning Research

Clea du Toit, Tran Quoc Bao Tran, Neha Deo, Sachin Aryal, Stefanie Lip, Robert Sykes, Ishan Manandhar, Aristeidis Sionakidis, Leah Stevenson, Harsha Pattnaik, Safaa Alsanosi, Maria Kassi, Ngoc Le, Maggie Rostron, Sarah Nichol, Alisha Aman, Faisal Nawaz, Dhruven Mehta, Ramakumar Tummala, Linsay McCallum, Sandeep Reddy, Shyam Visweswaran, Rahul Kashyap, Bina Joe and Sandosh Padmanabhan

Link to Paper

Sandosh Article image

Summary

Machine learning (ML) is a technology used to automate tasks and make complex decisions, and it's becoming more popular in hypertension research. Du Toit and an international team of collaborators analysed 63 articles on hypertension-related ML research and suggest that improvements in reporting quality and validation methods are needed to fully realize the potential of ML in hypertension care. Specifically, they highlight that careful attention should be paid to appropriate calibration and external validation of machine learning algorithms, as well as addressing algorithmic bias and inequity.


First published: 25 May 2023