Ovarian Cancer Detection using Menstruation and Weight/Calorie Tracking
The team’s chosen solution could increase the machine learning presence in medical diagnostics and reduce bias in medicine. However, if executed poorly, this project could result in user distress and potential legal implications due to personal injury law. It has little to no environmental impact.
The project team consists of four undergraduate computer engineering students with a project supervisor who is a Professor in the Department of Electrical and Computer Engineering.
The plan for this project is to automate symptom detection by developing a women’s health application. The main symptoms that are detected by the app are changes in menstruation and changes in diet. This solution eliminates the need for the patient’s initiative to seek diagnosis for their symptoms or even notice their symptoms. This application has two main features - a menstruation calendar and a calorie/diet tracker. Any significant differences in menstruation and diet are detected by the application and will redirect the user to fill out a survey with additional medical information. The survey and application data will be entered into a machine learning algorithm to determine whether the user should get tested for ovarian cancer.
Dr. Korenberg, P.Eng.
Nour Mahmoud
Naod Dereje
Brooke Henderson
Kavin Mohan