Application of Machine Learning in Radiation Oncology Scheduling

Cancer incidence rates in Canada are increasing steadily every year which puts a strain on the treatment system. In Quebec, the waiting time to start treatment of cancer patients is enforced by law, however, it is difficult to meet with limited resources and personnel. Efficient planning is hence vital in
reducing the long waiting time of cancer patients as well as improving resources’ utilization and wellness
of medical staff.
Gray is a company founded with the ambition of applying Operations Research and Machine
Learning solutions to improve healthcare quality. Their product GrayOS, which provides multiple
functionalities such as modelling patient flows, automatically scheduling cancer treatments, real-time
optimization, has been deployed and is in the testing phase at the cancer center of CHUM. In this project,
the intern is expected to work with Gray to develop a novel approach for scheduling radiotherapy
treatments.
Scheduling in healthcare is challenging due to the stochastic nature of the field. Future arrivals of
patients are not known in advance, yet have a huge impact on the current scheduling decision. Many
techniques have been developed to take into account future arrivals such as stochastic programming or
Markov Decision Process. However, those techniques usually are algorithmically heavy and might be
difficult to implement and maintain in real-world applications.

Faculty Supervisor:

Antoine Legrain

Student:

Partner:

Gray Oncology Solutions Inc.

Discipline:

Mathematics

Sector:

Professional, scientific and technical services

University:

Polytechnique Montréal

Program: