:param multi_channel_medical_img: shape of [batch, input_channels, height, width, depth] :return: late fusion classification predicti

:param multi_channel_medical_img: shape of [batch, input_channels, height, width, depth] :return: late fusion classification predicti

Returns the majority vote element of a list

Intensity normalization based on percentile Clips the range based on the quarile values. :param min_val: should be in the range [0,100]

Majority labeling vote to produce ground truth labels

Reads labels and returns them in a tuple of sorted lists

r""" Restores checkpoint from a pth file and restores optimizer state. Args: ckpt_file (str): A PyTorch pth file containi

r""" Saves checkpoint at a certain global step during training. Optimizer state is also saved together. Args: dir

Function to display a row of image slices Input is a list of numpy 2D image slices

Assuming dim 1 is the classes dim , it skips all the indexes after the desired class

Computes square angular loss between input and target directions. Makes sure that the input and target directions are normalized so that torc

Accepts nifty objects Transfers coordinate space from modality_2 to modality_1

this function will produce NON-overlaping sub-volumes prediction that produces full 3d medical image compare some slices with ground tru

Performs majority voting on the stacked labels

Args: img_numpy (numpy): Image to be flipped. label (numpy): Label segmentation map to be flipped Returns:

Args: img_numpy (numpy): Image to be flipped. label (numpy): Label segmentation map to be flipped Returns:

Args: img_numpy (numpy): Image to be rotated. label (numpy): Label segmentation map to be rotated Returns: