Automatically Assessing Frailty from Medical Notes

Frailty is an especially problematic expression of population ageing. It is a condition characterised by loss of biological reserves across multiple organ systems and vulnerability to physiological decompensation after a stressor event. Older people with frailty are at increased risk of adverse outcomes including disability, hospitalisation, nursing home admission and mortality. This project addresses the issue of predicting frailty in patients by automatically processing medical notes taken by health professionals when consulting those patients. The result of this research could provide valuable information for physicians and other health professionals for individual patient care, as well as epidemiologic indicators on the overall population. The problem will be addressed by evaluating three different, but maybe complementary, approaches: Natural Language Based (NLP) processing, frailty index prediction using deep learning and summarization techniques. Besides, we intend to study in this project two side aspects of the application of technology to the health domain: the explainability of automatically generated predictions and knowledge discovery.

Faculty Supervisor:

Randy Goebel

Student:

Juliano Cicero Bitu Rabelo;Housam Khalifa Bashier Babiker

Partner:

AltaML

Discipline:

Computer science

Sector:

Information and cultural industries

University:

University of Alberta