Task
Image Segmentation
C
ianpan/chest-x-ray-basic
Image Segmentation·ianpan· 2.6K· 9
This model performs both segmentation and classification on chest radiographs (X-rays). The model uses a tfefficientnetv2s backbone with a U-Net decoder for segmentation and linear layer for classification. For frontal radiographs, the model segments the: 1) right lung, 2) left lung, and 3) heart. The model also predicts the chest X-ray view (AP, PA, lateral), patient age, and patient sex. The Ch
# pull & run locally
pip install mlforge-sdk && mlforge pull ianpan/chest-x-ray-basic
pip install mlforge-sdk && mlforge pull ianpan/chest-x-ray-basic
Model details
Provider
ianpan
Framework
transformers
Parameters
22.2M
Size
170 MB
Downloads
2.6K
Likes
9
Updated
2025-01-03
About ianpan/chest-x-ray-basic
This model performs both segmentation and classification on chest radiographs (X-rays). The model uses a tfefficientnetv2s backbone with a U-Net decoder for segmentation and linear layer for classification. For frontal radiographs, the model segments the: 1) right lung, 2) left lung, and 3) heart. The model also predicts the chest X-ray view (AP, PA, lateral), patient age, and patient sex. The CheXpert (small version) and NIH Chest X-ray datasets were used to train the model. Segmentation masks were obtained from the CheXmask dataset (paper). The final dataset comprised 335,516 images from 96,385 patients and was split into 80% training/20% validation. A holdout test set was not used since minimal tuning was performed. The view classifier was trained only on CheXpert images (NIH images excluded from loss function), given that lateral radiographs are only present in CheXpert. This is to avoid unwanted bias in the model, which can occur if one class originates only from a single dataset.