Total processes on two-stage approach + corresponding ipynb notebook titles

1. preprocessing stage 1: windowing-preprocessing-technique-dataset.ipynb
2. stage 1 (multilabel): stage-1-window-multilabel.ipynb
3. preprocessing stage 2:
   • preprocess Yolov5: preprocess-cropping-vertebrae-with-yolo.ipynb
   • train Yolov5: train-infer-cropping-vertebrae-with-yolo.ipynb
   • Convert obtained Yolov5 output from jpg -> npz files. The output folders utilized in 'Cropping voxel': convert-to-npz-file-test-icon.ipynb
   • Cropping voxel, for processing vert_list.csv: stage2-preprocessing-sagittal-view-labels-only.ipynb
   • Cropping voxel: preprocessing-stage-2-cropping-voxel.ipynb
4. stage 2: stage-2-cracknet-lstm-yolo-window.ipynb

More visual process

please check the .pdf file attached for more visualzing methodology and total processes