Development of a clinical decision support system to augment multi-disciplinary tumor board decision-making

We propose to develop and evaluate a clinical decision support systems (CDSS) for the Princess Margaret Cancer Centre (PMCC) lung metastases multidisciplinary tumor board (MDT) to improve the consistency and quality of MDT decision-making. We will develop and train a machine learning-based (ML) classifier, using historical decisions data from lung cancer MDTs to model the decision-making process. Clinical parameters and lesion characteristics parameterized from CT imaging will be used as inputs to the ML-based classifier. The ML classifier will be one of several components in the CDSS Application, called Janus, including auto-segmentation of normal anatomy and lesions and contour parametrization tools. The proposed CDSS application will impact health care delivery at the partner institution in several ways including 1) improving lung metastases MDT efficiency, 2) providing quality assurance to ensure consistent, evidence-based decision making and 3) enable knowledge sharing by deploying the CDSS in other partner-institutions without similar MDTs.

Faculty Supervisor:

Dionne Aleman

Student:

Partner:

Princess Margaret Cancer Centre

Discipline:

Engineering

Sector:

Health and Related Sciences & Technology

University:

University of Toronto

Program: