Development and validation of a machine learning predictor for the early detection of prostate cancer

Prostate cancer is the third leading cause of death from cancer in men in Canada. However, prostate cancer is highly treatable if diagnosed early. Unfortunately, due to lack of cost-effective and meaning test detecting the early presence of the cancer. Most prostate cancer (92%) are found when the career is spreading to nearby organs. This project aims to address the gap by using machine learning methods to build a classifier for the early detection of prostate cancer which has higher accuracy and higher sensitivity than PSA testing (the current standard).Our partner organization, Metabolomics Technologies Inc. (MTI), has conducted a clinical trial and generated metabolomics data on Prostate cancer and health-control serum samples. If this test is more meaningful to currently used PSA test, it will increase the survival rate for prostate cancer.

Faculty Supervisor:

Michael Li

Student:

Zhengjun Liu

Partner:

Metabolomic Technologies Inc.

Discipline:

Mathematics

Sector:

Medical devices

University:

Program: