Identifying and classifying patient-uploaded image quality for teledermatology practices - QC-484

Project type: Research
Desired discipline(s): Computer science, Mathematical Sciences, Operations research
Company: ORO Health Inc / ORO Santé Inc
Project Length: Flexible
Preferred start date: As soon as possible.
Language requirement: Flexible
Location(s): Montreal, QC, Canada; Canada
No. of positions: 1
Desired education level: Undergraduate/BachelorMaster'sPhDPostdoctoral fellow
Search across Mitacs’ international networks - check this box if you’d also like to receive profiles of researchers based outside of Canada: 
No

About the company: 

ORO Health platform is a white-labeled, secure, user-and-doctor friendly platform that empowers other health practitioners to create their own virtual clinic with no coding or design skills needed.

Describe the project.: 

This project is about detecting the characteristics of the patient-uploaded picture, filtering, classifying to accept or reject them based on the clinic preference. The main goal of ORO is to develops a telemedicine platform that empowers both Doctor and Patient to use the best modern tools and technology available today.

The candidate will review the dataset classification, train models, compare results between models and approaches, work with our internal Data Science team to make new improvement hypotheses, test them out, present the result internally and externally.

The internal team is using iterative development and we enforce strong PEPIDA/HIPAA compliance so we follow the law's guidelines regarding patient data handling and protection. One important aspect is that project will not be concerned with making any diagnostic on the disease that may or may not be present in the patient pictures. The tool is a very important part of improving the efficiency and experience of teledermatology clinics.

Required expertise/skills: 

Machine Learning, Software (ideally Python), Image Classification algorithm and methods