Hierarchical Semi-Supervised Learning Framework for Surgical Gesture Segmentation and Recognition Based on Multi-Modality Data
This work addresses the problem of automating surgical workflow analysis for robot-assisted surgery, which is incremental as it builds on existing methods with multi-modality integration.
The authors tackled surgical gesture segmentation and recognition in robot-assisted surgery by developing a hierarchical semi-supervised learning framework using multi-modality data, achieving an average F1 score of 0.623 for segmentation and accuracy of 0.856 for recognition on the JIGSAWS database.
Segmenting and recognizing surgical operation trajectories into distinct, meaningful gestures is a critical preliminary step in surgical workflow analysis for robot-assisted surgery. This step is necessary for facilitating learning from demonstrations for autonomous robotic surgery, evaluating surgical skills, and so on. In this work, we develop a hierarchical semi-supervised learning framework for surgical gesture segmentation using multi-modality data (i.e. kinematics and vision data). More specifically, surgical tasks are initially segmented based on distance characteristics-based profiles and variance characteristics-based profiles constructed using kinematics data. Subsequently, a Transformer-based network with a pre-trained `ResNet-18' backbone is used to extract visual features from the surgical operation videos. By combining the potential segmentation points obtained from both modalities, we can determine the final segmentation points. Furthermore, gesture recognition can be implemented based on supervised learning. The proposed approach has been evaluated using data from the publicly available JIGSAWS database, including Suturing, Needle Passing, and Knot Tying tasks. The results reveal an average F1 score of 0.623 for segmentation and an accuracy of 0.856 for recognition.