Analasys of ML Methods for disease detection in tear residue

Inspiration

Our inspiration is to work on relevant medical research, especially related to AI

What it does

We fine-tuned a pre-trained CNN model to predict what disease a patient has from images of residue of dried tears. We also explored other prediction methods and compared them in our report

How we built it

Mostly python

Challenges we ran into

Poor dataset, hardware restrictions for training purposes. Analysis of the whole project scope, the data science part

Accomplishments that we're proud of

We did it

What we learned

We learned a lot about machine learning, classification, and about computer science application in the field of medicine

Built With

• python
• pytorch
• scikit-learn
• tensorflow