Automated detection of blood cell based diseases involving identifying and characterising patient blood samples using Machine learning and Deep Transfer Learning
- Each image has multiple blood cells in it. On average it would probably be 10 cells per image.
- The cell-of-interest is colored purple while other cells are colored skin-red.
- One cell-of-interest per image
- Goal: Identify the cell-type of cell-of-interest