Automated vertebral landmarks and spinal curvature estimation using non-directional part affinity fields

Acquire the datasets (see below).

Unzip the train, val, data into dataroot/boostnet_labeldata

Unzip the test data into dataroot/submit_test_images folder

Merge train and val csv annotation, put into dataroot/trainval-labels folder

By default, dataroot = ../ScoliosisData

Run resize_images.py to apply augmentation and resize

Run train.py to train on "train" set

Run eval.py to produce heatmaps

Run resize_images.py to flipLR, resize

Run train.py --trainval to train

Run eval.py --trainval to produce heatmaps

Run cobb_angle_eval.py to evaluate landmark pairs and Cobb angles

Dataset provided by:

Wu, Hongbo, et al. "Automatic landmark estimation for adolescent idiopathic scoliosis assessment using BoostNet." International Conference on Medical Image Computing and Computer-Assisted Intervention. Springer, Cham, 2017.