Most state-of-the-art methods for action segmentation are based on single input modalities or naïve fusion of multiple data sources. However, effective fusion of complementary information can potentially strengthen segmentation models and make them more robust to sensor noise and more accurate with smaller training datasets. In order to improve multimodal representation learning for action segmentation, we propose to disentangle hidden features of a multi-stream segmentation model into modality-shared components, containing common information across data sources, and private components; we then use an attention bottleneck to capture long-range temporal dependencies in the data while preserving disentanglement in consecutive processing layers. Evaluation on 50salads, Breakfast and RARP45 datasets shows that our multimodal approach outperforms different data fusion baselines on both multiview and multimodal data sources, obtaining competitive or better results compared with the state-of-the-art. Our model is also more robust to additive sensor noise and can achieve performance on par with strong video baselines even with less training data.

Automatically recognising surgical gestures from surgical data is an important building block of automated activity recognition and analytics, technical skill assessment, intra-operative assistance and eventually robotic automation. The complexity of articulated instrument trajectories and the inherent variability due to surgical style and patient anatomy make analysis and fine-grained segmentation of surgical motion patterns from robot kinematics alone very difficult. Surgical video provides crucial information from the surgical site with context for the kinematic data and the interaction between the instruments and tissue. Yet sensor fusion between the robot data and surgical video stream is non-trivial because the data have different frequency, dimensions and discriminative capability. In this paper, we integrate multimodal attention mechanisms in a two-stream temporal convolutional network to compute relevance scores and weight kinematic and visual feature representations dynamically in time, aiming to aid multimodal network training and achieve effective sensor fusion. We report the results of our system on the JIGSAWS benchmark dataset and on a new in vivo dataset of suturing segments from robotic prostatectomy procedures. Our results are promising and obtain multimodal prediction sequences with higher accuracy and better temporal structure than corresponding unimodal solutions. Visualization of attention scores also gives physically interpretable insights on network understanding of strengths and weaknesses of each sensor.

Surgical instrument segmentation can be used in a range of computer assisted interventions and automation in surgical robotics. While deep learning architectures have rapidly advanced the robustness and performance of segmentation models, most are still reliant on supervision and large quantities of labelled data. In this paper, we present a novel method for surgical image generation that can fuse robotic instrument simulation and recent domain adaptation techniques to synthesize artificial surgical images to train surgical instrument segmentation models. We integrate attention modules into well established image generation pipelines and propose a novel cost function to support supervision from simulation frames in model training. We provide an extensive evaluation of our method in terms of segmentation performance along with a validation study on image quality using evaluation metrics. Additionally, we release a novel segmentation dataset from real surgeries that will be shared for research purposes. Both binary and semantic segmentation have been considered, and we show the capability of our synthetic images to train segmentation models compared with the latest methods from the literature.

Surgical gesture recognition is important for surgical data science and computer-aided intervention. Even with robotic kinematic information, automatically segmenting surgical steps presents numerous challenges because surgical demonstrations are characterized by high variability in style, duration and order of actions. In order to extract discriminative features from the kinematic signals and boost recognition accuracy, we propose a multi-task recurrent neural network for simultaneous recognition of surgical gestures and estimation of a novel formulation of surgical task progress. To show the effectiveness of the presented approach, we evaluate its application on the JIGSAWS dataset, that is currently the only publicly available dataset for surgical gesture recognition featuring robot kinematic data. We demonstrate that recognition performance improves in multi-task frameworks with progress estimation without any additional manual labelling and training.

Kinematic trajectories recorded from surgical robots contain information about surgical gestures and potentially encode cues about surgeon's skill levels. Automatic segmentation of these trajectories into meaningful action units could help to develop new metrics for surgical skill assessment as well as to simplify surgical automation. State-of-the-art methods for action recognition relied on manual labelling of large datasets, which is time consuming and error prone. Unsupervised methods have been developed to overcome these limitations. However, they often rely on tedious parameter tuning and perform less well than supervised approaches, especially on data with high variability such as surgical trajectories. Hence, the potential of weak supervision could be to improve unsupervised learning while avoiding manual annotation of large datasets. In this paper, we used at a minimum one expert demonstration and its ground truth annotations to generate an appropriate initialization for a GMM-based algorithm for gesture recognition. We showed on real surgical demonstrations that the latter significantly outperforms standard task-agnostic initialization methods. We also demonstrated how to improve the recognition accuracy further by redefining the actions and optimising the inputs.