CatStep: Automated Cataract Surgical Phase Classification and Boundary Segmentation Leveraging Inflated 3D-Convolutional Neural Network Architectures and BigCat

Objective: Accurate identification of surgical phases during cataract surgery is essential for improving sur-gical feedback and performance analysis. Time spent in each surgical phase is an indicator of performance, and segmenting out specific phases for further analysis can simplify providing both qualitative and quantitative feedback on surgical maneuvers.Study Design: Retrospective surgical video analysis.Subjects: One hundred ninety cataract surgical videos from the BigCat dataset (comprising nearly 4 million frames, each labeled with 1 of 11 nonoverlapping surgical phases).Methods: Four machine learning architectures were developed for segmentation of surgical phases. Models were trained using cataract surgical videos from the BigCat dataset.Main Outcome Measures: Models were evaluated using metrics applied to frame-by-frame output and, uniquely in this work, metrics applied to phase output.Results: The final model, CatStep, a combination of a temporally sensitive model (Inflated 3D Densenet) and a spatially sensitive model (Densenet169), achieved an F1-score of 0.91 and area under the receiver operating characteristic curve of 0.95. Phase-level metrics showed considerable boundary segmentation performance with a median absolute error of phase start and end time of just 0.3 seconds and 0.1 seconds, respectively, a segmental F1-score @70 of 0.94, an oversegmentation score of 0.89, and a segmental edit score of 0.92.Conclusion: This study demonstrates the feasibility of high-performance automated surgical phase identi-fication for cataract surgery and highlights the potential for improved surgical feedback and performance analysis.Financial Disclosure(s): Proprietary or commercial disclosure may be found in the Footnotes and Disclo-sures at the end of this article. Ophthalmology Science 2024;4:100405 (c) 2023 by the American Academy of Ophthalmology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).

Automated summary

Accurate identification of surgical phases is crucial for improving surgical feedback and performance analysis. This study developed four machine learning architectures and achieved high-performance automated phase identification for cataract surgery, providing both qualitative and quantitative feedback and highlighting its potential.