Low cost solution to detect chest pathologies deployed to Raspberry Pi and Flask without radiologists
https://www.youtube.com/watch?v=QC-d8nf_SdA&feature=youtu.be
CheXpert is a large dataset of chest X-rays and competition for automated chest x-ray interpretation, which features uncertainty labels and radiologist-labeled reference standard evaluation sets. CheXpert is a large public dataset for chest radiograph interpretation, consisting of 224,316 chest radiographs of 65,240 patients. https://stanfordmlgroup.github.io/competitions/chexpert/
Here is a nice explanation by Qin (Erin) Hu regarding the X-Pi Model.
https://www.youtube.com/watch?v=DH-LfOf_ia4&feature=youtu.be