Using weakly supervised text classification on patient free text comments

Abstract: Free text comments (FTC) of patient-reported outcome measures (PROMS) offer invaluable insight into health-related quality of life (HRQoL). However, extracting meaningful information from FTC is challenging due to the time-consuming nature of manual analysis methods (the common analysis method). However, weakly supervised text classification (WSTC) can be a valuable method of analysis to classify domain-specific text data in which there is limited labelled data. I will present the use of weakly supervised text classification (WSTC) method to analyse PROMs FTC data. Specifically, I apply five keywords-based WSTC methods to FTC PROMs data from two national PROMs datasets: prostate cancer and colorectal cancer. The presentation will cover key findings, including how effectively these methods can identify and categorise HRQoL themes in the FTCs.

Bio: Anna-Grace Linton is a PhD student in the CDT in AI for Medical Diagnosis and Care, University of Leeds. Her research focuses on employing Natural language Processing (NLP) techniques to analyse free text comments within PROMs, for a greater understanding of the HRQoL of cancer patients.