Automated Operative Phase and Step Recognition in Vestibular Schwannoma Surgery: Development and Preclinical Evaluation of a Deep Learning Neural Network (IDEAL Stage 0)
Authors:
- Simon C. WilliamsSWSimon C. Williams
- Dorothée DuvauxDDDorothée Duvaux
- Adrito DasADAdrito Das
- Siddharth SinhaSSSiddharth Sinha
- Hugo Layard HorsfallHHHugo Layard Horsfall
- Jonathan P. FunnellJFJonathan P. Funnell
- John G. HanrahanJHJohn G. Hanrahan
- Danyal Z. KhanDKDanyal Z. Khan
- William MuirheadWMWilliam Muirhead
- Neil KitchenNKNeil Kitchen
- Francisco VasconcelosFVFrancisco Vasconcelos
- Sophia BanoSBSophia Bano
Open Access
- 30 April 2025
- journal article
- Published by Wolters Kluwer Health in Neurosurgery
- Vol. 98 (4) , 799-809
Abstract
BACKGROUND AND OBJECTIVES: Machine learning (ML) in surgical video analysis offers promising prospects for training and decision support in surgery. The past decade has seen key advances in ML-based operative workflow analysis, though existing applications mostly feature shorter surgeries (5 hours duration), data-heavy surgeries, using VS resection as an exemplar. METHODS: A video dataset of 21 microscopic retrosigmoid VS resections was collected at a single institution over 3 years and underwent workflow annotation according to a previously agreed expert consensus (Approach, Excision, and Closure phases; and Debulking or Dissection steps within the Excision phase). Annotations were used to train a ML model consisting of a convolutional neural network and a recurrent neural network. 5-fold cross-validation was used, and performance metrics (accuracy, precision, recall, F1 score) were assessed for phase and step prediction. RESULTS: Median operative video time was 5 hours 18 minutes (IQR 3 hours 21 minutes-6 hours 1 minute). The “Tumor Excision” phase accounted for the majority of each case (median 4 hours 23 minutes), whereas “Approach and Exposure” (28 minutes) and “Closure” (17 minutes) comprised shorter phases. The ML model accurately predicted operative phases (accuracy 81%, weighted F1 0.83) and dichotomized steps (accuracy 86%, weighted F1 0.86). CONCLUSION: This study demonstrates that our ML model can accurately predict the surgical phases and intraphase steps in retrosigmoid VS resection. This demonstrates the successful application of ML in operative workflow recognition on low-volume, lengthy, data-heavy surgical videos. Despite this, there remains room for improvement in individual step classification. Future applications of ML in low-volume high-complexity operations should prioritize collaborative video sharing to overcome barriers to clinical translation.Keywords
This publication has 49 references indexed in Scilit:
- Association Between Facility Surgical Volume and Excess Time in the Hospital After Surgical Resection of Vestibular SchwannomasJAMA Otolaryngology–Head & Neck Surgery, 2023
- TEsoNet: knowledge transfer in surgical phase recognition from laparoscopic sleeve gastrectomy to the laparoscopic part of Ivor–Lewis esophagectomySurgical Endoscopy, 2023
- Evolution of the digital operating room: the place of video technology in surgeryLangenbeck's Archives of Surgery, 2023
- Artificial intelligence prediction of cholecystectomy operative course from automated identification of gallbladder inflammationSurgical Endoscopy, 2022
- SAGES consensus recommendations on an annotation framework for surgical videoSurgical Endoscopy, 2021
- Real-time automatic surgical phase recognition in laparoscopic sigmoidectomy using the convolutional neural network-based deep learning approachSurgical Endoscopy, 2019
- Assessment of Automated Identification of Phases in Videos of Cataract Surgery Using Machine Learning and Deep Learning TechniquesJAMA Network Open, 2019
- Observational clinical human reliability analysis (OCHRA) for competency assessment in laparoscopic colorectal surgery at the specialist levelSurgical Endoscopy, 2011
- Constructing hierarchical task analysis in surgerySurgical Endoscopy, 2007
- Bestimmung der Effizienz von intraoperativer TechnologieHNO, 2006