Dataset: The “input" contains 45 cases of multi-sequence CMR. Each case refers to a patient with three sequence CMR, i.e., LGE, T2 and bSSFP CMR. The “masks” folder contains 45 cases of multi-sequence CMR, where each mask represents the segmentation map of the the corresponding CMR Slice image (e.g image: myops_training_108_CO_0.png mask:myops_training_108_C0_gd_0.png ).

• Label 1: normal myocardium
• Label 2: edema
• Label 3: scar
• Label 0: Background

Dataloader: The data loader can be set to use online augmentation in order to overcome the limited data and overfitting problems. The augmentation is done with albumentation library and can include:

• Rotation: one from Rotate, VerticalFlip, HorizontalFlip, RandomRotate90, Transpose or ShiftScaleRotate
• Cropping: CenterCrop or RandomCrop
• Contrast Enhancement: RandomBrightnessContrast, RandomGamma or CLAHE (Contrast Limited Adaptive Histogram Equalization)
• Noise Addtion: Gaussian Noise
• Distrorsion: GridDistortion or Elastic Transform
• Blurring: One from Blur, MotionBlur or GaussianBlur

An Example of augmented Image and overlayed mask can be seen here:

How to train the model: To train the model, please run the following command, you can change the parameters within the train.py file.

python -u src\trainer.py -nc 3 -e 150

*** Default training parameteres ***
parser.add_argument("-lr", help="set the learning rate for the unet", type=float, default=0.0001)
parser.add_argument("-e", "--epochs", help="the number of epochs to train", type=int, default=300)
parser.add_argument("-gn", "--gaussianNoise", help="whether to apply gaussian noise", action="store_true",
                    default=True)
parser.add_argument("-gpu",  help="Det the device to use the GPU", type=bool, default=False)
parser.add_argument("--n_samples", help="number of samples to train", type=int, default=100)
parser.add_argument("-bs", "--batch_size", help="batch size of training", type=int, default=4)
parser.add_argument("-nc", "--n_class", help="number of classes to segment", type=int, default=3)
parser.add_argument("-nf", "--n_filter", help="number of initial filters for Unet", type=int, default=32)
parser.add_argument("-nb", "--n_block", help="number unet blocks", type=int, default=4)
parser.add_argument("-pt", "--pretrained", help="whether to train from scratch or resume", action="store_true",
                    default=False)

• What, if anything, did you do to verify that the segmentation masks and images were correctly aligned in the data loader?

  • You could easily check the image dimension after plotting the first set of images to verify if they are aligned or not.
    • Debug manually step by step to check dimensions and range
    • Intermediate plotting of the results.

• What assumptions did you make about the data or model training during this process? *