Omar Dervisevic - Dr. Bostrom - Week 6

 This week was one focused more in the lab. I was finalizing the annotations for the slides that I had scanned to train an AI model. I met with Dr. Richard Bell, a post-doc in the lab, and we spoke about what the next steps would be. The first step would be to attempt to modify the current model. This is the first step because there are tissue types that the original model was able to distinguish very well so attempting to train the model on the other tissue types could work. Another approach would be to create a whole new model that will only segments the tissues that I have annotated and run it in parallel with the current model and only make the outputs the tissues i want from the better models. This would allow for the current best of both models to be incorporated in the final result. With the two possible approach we will try them the following week and see which one works better. After this is done we will start to implement the grading of the cartilage damage. With these steps laid out it will make it so that we are closer to the final goal of the project which is to automate the grading of the cartilage damage.