Student Project in Collaboration with Feinberg School of Medicine Published in the Journal of Pain

MSiA students help to improve medical pain classification tools in "Everything Starts with Data" course

MSiA students used a machine learning model to extract markings from the Adolescent Pediatric Pain Tool (APPT) diagrams and identify corresponding body locations for patients undergoing a minimally invasive thoracoscopic procedure to repair pectus excavatum (Nuss procedure) and spinal fusion.

Correct classification rate for each body location per individual patient was 97.2%. Overall sensitivity was 95.5% and specificity 97.3%. This computer vision method of coding body outline diagrams is accessible, convenient, and provides better sensitivity and specificity for pain site and surface area than manual coding.

This study was conducted under the guidance of Dr. Renee Manworren from Feinberg School of Medicine as a part of the MSiA course “Everything Starts with Data” in the fall quarter of 2020. The abstract of this study was published in the recent issue of Journal of Pain.

McCormick News Article