I'm working on a project to detect an illness from ECG images but I have a small number of data (250 images) so the model is overfitting.
I tried many different things but still no progress. it looks like the augmentation of data is the last option but the rotation of images won't be very helpful since we are working with medical data. I'm looking if there is any possibility to transform the images into serial data and maybe do the augmentation after that, if yes do you have any idea how to do it. Thanks for taking the time to read my post!