Abstract: The effectiveness and robustness of medical image segmentation algorithms inextricably depends on the quality and quantity of training data. Insufficient data continually restricts model ...