Medical errors, particularly missed diagnoses, remain a leading cause of preventable harm in healthcare (Makary & Daniel, 2016). This project proposes a novel, scalable diagnostic paradigm in which individual laboratory tests are trained to predict their own abnormality using real-world patient data. Through exploratory data analysis, natural language processing with clinical text embeddings, and multiple machine learning models, we predict lab values from patient history, vitals, and related tests. This reverse, test centric approach reduces reliance on subjective diagnoses, mitigates labeling bias, and enables objective, incremental deployment of AI driven diagnostics to improve early detection and clinical decision-making.
Watch the team present this project at 53:15 in the session recording here.
Keywords: medical errors, predictive modeling, healthcare, natural language processing, clinical embeddings, machine learning
Faculty Advisor
Ashish Pujari is a Data and AI consultant, practitioner, and educator with over 20 years of experience in in machine learning, big data, and cloud computing. He has led large global technology and data science teams and consulted for Fortune 500 companies in banking, finance, healthcare, insurance, and manufacturing.
As a Principal ML/AI Architect at AWS, Ashish provides strategic guidance to enterprise customers on leveraging the cloud for AI and Machine Learning. Prior to joining Amazon, he served in various technology leadership roles at Credera, GLG, IRI, and Pegasystems. Ashish holds a Master of Science in Analytics from the University of Chicago and BS in Electrical Engineering from the National Institute of Technology, Rourkela.