You may be interested in the just published review paper on deep learning enabled medical computer vision.
\nIt's not a technical ML paper (I think it's more for a medical audience) so don't expect to necessarily find cool new modelling techniques that will help you in this competition - although the cited articles are more \"hands-on\". However, it's an interesting read and I did not know that there are \"at-home smart toilets outfitted with diagnostic CNNs on cameras\"?!?!!
\nPerhaps more usefully for this competition, there's a reminder that there's \"1 million annotated, open-source images\" for chest x-rays in MIMIC-CXR, CheXpert and ChestX-ray8. That could of course be pretty useful for things like learning representations or semi-supervised approaches.
", "messages": [ { "id": 1145888, "postDate": "2021-01-09T12:12:33.100Z", "content": "You may be interested in the just published review paper on deep learning enabled medical computer vision.
\nIt's not a technical ML paper (I think it's more for a medical audience) so don't expect to necessarily find cool new modelling techniques that will help you in this competition - although the cited articles are more \"hands-on\". However, it's an interesting read and I did not know that there are \"at-home smart toilets outfitted with diagnostic CNNs on cameras\"?!?!!
\nPerhaps more usefully for this competition, there's a reminder that there's \"1 million annotated, open-source images\" for chest x-rays in MIMIC-CXR, CheXpert and ChestX-ray8. That could of course be pretty useful for things like learning representations or semi-supervised approaches.
", "rawMarkdown": "You may be interested in the just published review paper on [deep learning enabled medical computer vision](https://doi.org/10.1038/s41746-020-00376-2).\n\nIt's not a technical ML paper (I think it's more for a medical audience) so don't expect to necessarily find cool new modelling techniques that will help you in this competition - although the cited articles are more \"hands-on\". However, it's an interesting read and I did not know that there are *\"[at-home smart toilets](https://doi.org/10.1038/s41551-020-0534-9) outfitted with diagnostic CNNs on cameras\"*?!?!!\n\nPerhaps more usefully for this competition, there's a reminder that there's \"1 million annotated, open-source images\" for chest x-rays in [MIMIC-CXR](https://doi.org/10.1038/s41597-019-0322-0), [CheXpert](https://arxiv.org/abs/1901.07031) and [ChestX-ray8](https://arxiv.org/abs/1705.02315). That could of course be pretty useful for things like learning representations or semi-supervised approaches.", "votes": 1 } ], "comments": [], "raw_markdown_by_id": { "1145888": "You may be interested in the just published review paper on [deep learning enabled medical computer vision](https://doi.org/10.1038/s41746-020-00376-2).\n\nIt's not a technical ML paper (I think it's more for a medical audience) so don't expect to necessarily find cool new modelling techniques that will help you in this competition - although the cited articles are more \"hands-on\". However, it's an interesting read and I did not know that there are *\"[at-home smart toilets](https://doi.org/10.1038/s41551-020-0534-9) outfitted with diagnostic CNNs on cameras\"*?!?!!\n\nPerhaps more usefully for this competition, there's a reminder that there's \"1 million annotated, open-source images\" for chest x-rays in [MIMIC-CXR](https://doi.org/10.1038/s41597-019-0322-0), [CheXpert](https://arxiv.org/abs/1901.07031) and [ChestX-ray8](https://arxiv.org/abs/1705.02315). That could of course be pretty useful for things like learning representations or semi-supervised approaches." } }