Monitoring tool usage in surgery videos using boosted convolutional and recurrent neural networks

TL;DR

This paper presents a novel boosting strategy for training convolutional and recurrent neural networks simultaneously to improve automatic tool usage monitoring in surgical videos, achieving high accuracy in real-time and offline modes.

Contribution

It introduces a new boosting approach that trains CNNs and RNNs together for better feature extraction in surgical tool analysis, addressing the limitations of separate training.

Findings

Achieved high classification accuracy with A_z > 0.99 in both datasets.

Demonstrated effectiveness in both offline and online modes.

Validated approach on cataract and cholecystectomy videos.

Abstract

This paper investigates the automatic monitoring of tool usage during a surgery, with potential applications in report generation, surgical training and real-time decision support. Two surgeries are considered: cataract surgery, the most common surgical procedure, and cholecystectomy, one of the most common digestive surgeries. Tool usage is monitored in videos recorded either through a microscope (cataract surgery) or an endoscope (cholecystectomy). Following state-of-the-art video analysis solutions, each frame of the video is analyzed by convolutional neural networks (CNNs) whose outputs are fed to recurrent neural networks (RNNs) in order to take temporal relationships between events into account. Novelty lies in the way those CNNs and RNNs are trained. Computational complexity prevents the end-to-end training of "CNN+RNN" systems. Therefore, CNNs are usually trained first, independently from the RNNs. This approach is clearly suboptimal for surgical tool analysis: many tools are very similar to one another, but they can generally be differentiated based on past events. CNNs should be trained to extract the most useful visual features in combination with the temporal context. A novel boosting strategy is proposed to achieve this goal: the CNN and RNN parts of the system are simultaneously enriched by progressively adding weak classifiers (either CNNs or RNNs) trained to improve the overall classification accuracy. Experiments were performed in a dataset of 50 cataract surgery videos and a dataset of 80 cholecystectomy videos. Very good classification performance are achieved in both datasets: tool usage could be labeled with an average area under the ROC curve of $A_z = 0.9961$ and $A_z = 0.9939$, respectively, in offline mode (using past, present and future information), and $A_z = 0.9957$ and $A_z = 0.9936$, respectively, in online mode (using past and present information only).