27 ms·
This paper covers many demographics, not just ethnicity. So you are wanting to really do "model = models.get(a, b, x, y, z)". It may be possible, but with medic
by teruakohatu 2y ago
This paper covers many demographics, not just ethnicity. So you are wanting to really do "model = models.get(a, b, x, y, z)". It may be possible, but with medical data in such short supply, you would be lucky to get a large enough dataset.
You would also need to assume there no overlap in the distribution in the training data for each strata of training data. I would imagine that if you taught someone to detect cancer in x-rays, and only used images of people aged 80+, you might still have some success (>0.5) at detecting cancer in people aged 20-30.
- sdenton4 2y agoThis seems like a place where generative augmentation could help - generate an image of a person with demographic values x, y, z and an early stage lung cancer. Then you can generate images for more corners of the days, more evenly, and train the classifier on less biased data.
- teruakohatu 2y agoIf your model already understand the distribution of the data (fundamentally this is what statistical models are trying to discover) well enough for generative models then they probably can do classification well enough. The greater issue is that, for privacy and other reasons, there are many demographics that are not represented at all, or under represented, in medical image datasets.
- CoastalCoder 2y agoJust in case you'd want to know, the singular of "strata" is "stratum". (Although apparently that's sometimes debated [0].) [0] https://www.dictionary.com/browse/stratum https://www.dictionary.com/browse/stratum