Text this: Transformer and group parallel axial attention co-encoder for medical image segmentation.