There\u2019s more of like two parts with this question, so yeah. <\/p>\n
First question: Let\u2019s say I want to train a ML model to detect a basic disease based off an image, say a brain. I can find a large dataset on regular. Then, I find multiple smaller datasets with not as many brain with disease images. Thus, I take all these smaller datasets of brains with diseases, combine them into one, then use this new dataset (brain with diseases) and the other dataset (large dataset with regular brain), and use them for classification. Is this possible?<\/p>\n
Second question: can we extend this to multiple classes? Say we have a disease that requires many conditions\/symptoms to detect. Can I find these conditions from multiple data sets (One dataset contains characteristics, one dataset contains duration, one dataset includes images, etc) and essentially merge them all into one as long as they classify the same disease??<\/p>\n<\/div>\n
submitted by \/u\/ResearchingTinBot <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
