Machine Learning and Artificial Intelligence in Healthcare: Infectious Diseases Case Study and Broader Implications
Overview
Abstract
Dr. Ljubomir Buturovic is VP Machine Learning at Inflammatix, Inc.,
Sunnyvale, CA. He has been leading development of ML-powered clinical diagnostic tests since 2004. His track record includes two FDA-cleared clinical tests for diagnosing cancer of unknown primary, and an affordable breast cancer prognostic test used in low-and-middle-income countries. Dr. Buturovic published 70 manuscripts and is co-inventor on six issued US patents. He has been leading the development of TriVerity algorithms for the diagnosis and prognosis of infectious diseases and sepsis since 2017.
Speaker Biography
The use of ML/AI in healthcare is expanding rapidly. We discuss a case study in Emergency Room. Inflammatix developed a rapid Point-Of-Care Myrna instrument and a TriVerity diagnostic test for detection and prognosis of suspected acute infections and sepsis in Emergency Room. Myrna measures abundances of 29 messenger RNAs in patient’s blood, subsequently used as features for ML. The classifiers convert the input features to an intuitive test report comprising the separate likelihoods of (1) a bacterial infection (2) a viral infection, and (3) disease severity. The Myrna, TriVerity system, in development for 7 years, was granted Breakthrough Device designation by the United States Food and Drug Administration (FDA) and is currently under review for marketing clearance. We describe the challenges of translating academic research concept to an ML-powered clinical product aimed at improving patient care, and discuss broader lessons for the AI/ML applications in healthcare.