Exploring machine-learning and quantum machine-learning DFT-fueled approaches to materials discovery for medical imaging.

Medical imaging techniques, such as X-rays and CT scans, are vital for the accurately detecting diseases. However, medical imagining techniques subject the patients to ionizing radiation, with the cumulative effect of these repeated imaging procedures over one’s lifetime leading to higher risk of cardiovascular disease and cancer. Children are more vulnerable to radiation effects. To address this concern, advanced detector materials that can capture high quality images at a significantly lower radiation dosage are required. This study is focused on understanding and designing safe and non-toxic materials named perovskites that have shown promise in imaging devices of the future. To design enhanced perovskite materials, this project will be applying multi-scale computational techniques to understand the chemical behavior of these materials and studying their X-ray detection ability, longevity, and degradation. Following, advanced quantum machine-learning techniques will be applied to discover the most suitable materials in a faster way than through the classical trial-and-error method. The most promising perovskite candidates found from this project can be used to develop enhanced medical imagining devices that allows doctors to diagnose with greater accuracy, and at the same time, reducing radiation dose to the patient.

Faculty Supervisor:

Dennis Salahub

Student:

Partner:

Université de Rennes 1

Discipline:

Physics

Sector:

Education

University:

University of Calgary

Program: