SPaDe: A Synonym-based Pain-level Detection Tool for Osteoarthritis

Yuhao Chen, Boqun Shu, Mojtaba Moattari, Farhana H. Zulkernine, John A. Queenan, David Barber

Abstract: Osteoarthritis (OA) is a progressive chronic joint disease resulting in a breakdown of articular cartilage and bone when damaged joint tissues are not able to normally repair themselves. Over 300 million adults around the world suffer from symptomatic OA. The aim of this pilot research project is to understand the pain severity for OA and apply Information Extraction, Natural Language Processing and Machine Learning techniques to patients’ primary care electronic medical records and physicians’ chart notes and identify patients suffering from moderate-to-severe pain due to osteoarthritis. Pain level is explained using a variety of words in the chart notes and poses a difficult challenge for pain categorization only based on the words. We propose a pain level detection pipeline that integrates a synonym-based pain-level detection algorithm and unsupervised pain level detector. The results demonstrate the potential of our approach to identify patients with moderate to severe pain.