Leveraging Transfer Learning with Class-Specific Decoders for Laparoscopic Segmentation

Effective multi-organ segmentation in surgical data requires learning the intricate anatomical features and alleviating the challenge of class imbalance, which results from relatively lower proportions of small and limitedly exposed structures. Recent works on laparoscopic multi-organ segmentation focus on learning structure-specific features through class-specific decoder architectures and report favorable results. This work aims to extend the decoder-focused architectures to investigate knowledge sharing in encoded features, particularly in knowledge transfer across datasets. Additionally, we compare the feature adaptation for the encoder and decoder at different training stages. Besides corroborating previous findings on decoder-specific architectures, our results exhibit that transfer learning enabled faster training convergence and superior segmentation performance.