SV-RCNet: Workflow Recognition From Surgical Videos Using Recurrent Convolutional Network
The DOI matches its Crossref record; 48 of the 50 references it cites were found in a registry or an index.
Abstract
We propose an analysis of surgical videos that is based on a novel recurrent convolutional network (SV-RCNet), specifically for automatic workflow recognition from surgical videos online, which is a key component for developing the context-aware computer-assisted intervention systems. Different from previous methods which harness visual and temporal information separately, the proposed SV-RCNet seamlessly integrates a convolutional neural network (CNN) and a recurrent neural network (RNN) to form a novel recurrent convolutional architecture in order to take full advantages of the complementary information of visual and temporal features learned from surgical videos. We effectively train the SV-RCNet in an end-to-end manner so that the visual representations and sequential dynamics can be jointly optimized in the learning process. In order to produce more discriminative spatio-temporal features, we exploit a deep residual network (ResNet) and a long short term memory (LSTM) network, to extract visual features and temporal dependencies, respectively, and integrate them into the SV-RCNet. Moreover, based on the phase transition-sensitive predictions from the SV-RCNet, we propose a simple yet effective inference scheme, namely the prior knowledge inference (PKI), by leveraging the natural characteristic of surgical video. Such a strategy further improves the consistency of results and largely boosts the recognition performance. Extensive experiments have been conducted with the MICCAI 2016 Modeling and Monitoring of Computer Assisted Interventions Workflow Challenge dataset and Cholec80 dataset to validate SV-RCNet. Our approach not only achieves superior performance on these two datasets but also outperforms the state-of-the-art methods by a significant margin.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Another line of work uses CV to recognize distinct phases of surgery during operations, towards developing context-aware computer assistance systems 126,127 .”
From Deep learning-enabled medical computer vision (2021) · cites this paper as [126]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 4×
“Specifically, we exploit the very deep residual network, the ResNet [21], to extract highly discriminative visual features from the video frames.”
BackgroundMethodInfluentialGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
A survey on deep learning in medical image analysis
Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., van der Laak, J. A. W. M., van Ginneken, B., & Sánchez, C. I. (2017) · Medical Image Analysis · doi:10.1016/j.media.2017.07.005 · Cited by 15,332
Learning Deep Features for Discriminative Localization
Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., & Torralba, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.319 · Cited by 11,428 · cited 1×
“networks on an image and intercept the learned information of networks [37], [38].”
BackgroundCaffe: Convolutional Architecture for Fast Feature Embedding
Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S., & Darrell, T. (2014) · Proceedings of the 22nd ACM international conference on Multimedia · doi:10.1145/2647868.2654889 · Cited by 11,077 · cited 1×
“Our framework is implemented with C++ and Python based on the Caffe [35] deep learning library, using a TITAN X GPU for acceleration.”
MethodShin, H.-C., Roth, H. R., Gao, M., Lu, L., Xu, Z., Nogues, I., Yao, J., Mollura, D., & Summers, R. M. (2016) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2016.2528162 · Cited by 5,759
Long-term recurrent convolutional networks for visual recognition and description
Donahue, J., Hendricks, L. A., Guadarrama, S., Rohrbach, M., Venugopalan, S., Darrell, T., & Saenko, K. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298878 · Cited by 5,228 · cited 2×
“HMM, we propose to tap into the temporal dimension of the surgical video data using the LSTM [31], [32], which has been demonstrated as a very powerful tool to model temporal concepts.”
MethodDeep learning in medical image analysis
Shen, D., Wu, G., & Suk, H.-I. (2017) · Annual Review of Biomedical Engineering · doi:10.1146/annurev-bioeng-071516-044442 · Cited by 4,968
Deep visual-semantic alignments for generating image descriptions
Karpathy, A., & Fei-Fei, L. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298932 · Cited by 4,902
Generating Sequences With Recurrent Neural Networks
Graves, A. (2013) · arXiv · doi:10.48550/arxiv.1308.0850 · Cited by 2,894 · cited 1×
“HMM, we propose to tap into the temporal dimension of the surgical video data using the LSTM [31], [32], which has been demonstrated as a very powerful tool to model temporal concepts.”
Method
Research works printed but not linked: 2
- olah, 2015, Understanding LSTM networks, Proc GitHub Blog
- dergachyova, 0, Data-driven surgical workflow detection: Technical report for M2CAI 2016 surgical workflow challenge, M2CAI Challenge
Cited by
Items 301–316 of 316
Stabilizing Temporal Inference Dynamics for Online Surgical Phase Recognition
Liu, Y., Zhu, N., Peng, J., Chen, X., Granados, A., Wang, G., & Sebastien Ourselin, A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38233-7_49 · Cited by 1
Chen, Y.-Y., Wang, K.-N., Tayupo, D., Huaulmé, A., Timoh, K. N., Jannin, P., & Dou, Q. (2025) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-026-03642-y · Cited by 1
“With temporal extensions, SV-RCNet [5] and TeCNO [6] boost accuracy to 75.88% and 76.25%, yielding clear improvements over the single-frame baseline.”
InfluentialSelf-citationDV-Net: Explainable model for hemorrhage prediction in AIS patients via physician logic
Lan, X., Cui, S., Xu, H., Huang, Y., Wang, Y., Ren, H., Li, Y., & Jiang, J. (2026) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2026.110627
Please follow the rules: surgical workflow recognition constrained by linear temporal logic
Tayupo, D., Huaulmé, A., Timoh, K. N., Baxter, J. S. H., & Jannin, P. (2026) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-026-03724-x
“Furthermore, the use of temporal sequence models such as LSTMs [10], multistage temporal convolutional networks (MS-TCNs) [3], or transformers [5] has also permitted a better understanding of the sequence of actions performed during the surgery.”
Shen, D., Yang, S.-J., Low, C., Li, Q., Xu, M.-Y., Dou, Q., & Jin, Y. (2026) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2026.3699201
“Due to the cable-driven mechanism of RMIS robots [8], [9], pose labels derived from forward kinematics are unreliable [10], [11].”
Self-citationSugimori, H., Sugiyama, T., Poudel, S., Togo, R., Tang, M.-H., Feng, H., Koyano, H., Ujihara, M., Hirata, K., Ogawa, T., et al. (2026) · Applied Sciences · doi:10.3390/app16147118
Mahmood, M. T., Saadi, M. A. M. A., Qasim, M. H., Alkhafaji, S. M. S., Basil, M. A., Baker, A. S., & Ahmed, S. R. (2026) · International Symposium on AI-Driven Engineering Systems (ISADES) · doi:10.1109/isades69945.2026.11608185
Qu, B., Xiao, Y., Liu, W., Zheng, W., Wang, Z., Zhong, X., & Sheng, B. (2026) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2026.3716009
“EndoNet [9] extracted spatial features for tool presence and procedural stage classification; SV-RCNet [21] combined ResNet backbones with LSTMs to capture temporal dynamics; and TeCNO [22] employed dilated temporal convolutional networks to model long-range dependencies.”
MURPHY: improving surgical workflow analysis with large-scale datasets and relation-aware models
Zhao, S., Liu, Y., Wang, Q., Sun, D., Liu, R., & Zhou, S. K. (2026) · npj Digital Surgery · doi:10.1038/s44484-026-00011-9
Xie, H., Chen, X., Law, N. F., Chan, Y., Lam, K., Li, K. K. W., Lai, T. H. T., & Chu, V. (2026) · Proceedings - International Conference on Image Processing · doi:10.1109/icip61757.2026.11630341
Mohamed, S., Gammulle, H., Fernando, T., Sridharan, S., Lucey, S., & Fookes, C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38233-7_35
Test-Time Adaptation for Rare Surgical Phase Recognition: Bridging the Coverage-Gap Paradox
Park, H.-M., Tozzi, F., Muynck, R. D., Kim, N., Rashidian, N., Willaert, W., Neve, W. D., & Vankerschaver, J. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38233-7_57
Extended KAFR: a kinematic-adaptive paradigm for the efficient analysis of surgical video
Phong, N. H., Khairnar, S. M., & Sankaranarayanan, G. (2026) · npj Digital Surgery · doi:10.1038/s44484-026-00022-6
Farzadmanesh, M. J., Jaliliyan, A., Tavana, P., & Pazouki, A. (2026) · Surgical Endoscopy · doi:10.1007/s00464-026-13404-1
Mank, Q. J., Korsten, T., Doornbos, M.-C. J., Li, Y., Maat, A. P. W. M., Siregar, S., Suyker, W. J. L., Kluin, J., Khalil, Y. A., & Sadeghi, A. H. (2026) · Innovations Technology and Techniques in Cardiothoracic and Vascular Surgery · doi:10.1177/15569845261487190
Agrawal, N., & Kharat, M. (2026) · BMC Artificial Intelligence · doi:10.1186/s44398-026-00038-5
“Surgical procedures exhibit highly structured sequential patterns with characteristic phase transitions, yet individual phase durations vary substantially based on patient anatomy, pathology, and surgeon technique [6].”
Metadata
- Authors
- Jin, Y., Dou, Q., Chen, H., Yu, L., Qin, J., Fu, C.-W., & Heng, P.-A.
- Affiliations
- Chinese University of Hong Kong (HK)
- Hong Kong Polytechnic University (HK)
- Venue
- IEEE Transactions on Medical Imaging
- Published
- 2018-05
- Volume
- 37
- Issue
- 5
- Pages
- 1114-1126
- ISSN
- 0278-0062 (ISSN-L)
- 1558-254X
- OpenAlex
- W2777273430
- Cited by
- 316
- References
- 50
- Citations by year
- 2026: 32 · 2025: 55 · 2024: 37 · 2023: 45 · 2022: 45 · 2021: 50 · 2020: 30 · 2019: 17 · 2018: 4
- FWCI
- 14.22
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Transactions on Medical Imaging (publishedVersion)
- PubMed (publishedVersion)
- Rare & Special e-Zone (The Hong Kong University of Science and Technology) (submittedVersion)
- PolyU Institutional Research Archive (Hong Kong Polytechnic University) (submittedVersion)
- Funders
- Research Grants Council of Hong Kong Special Administrative Region (CUHK 14202514, CUHK 14203115)
- Shenzhen Science and Technology Program (JCYJ20170413162617606)
- Hong Kong Polytechnic University (1-ZE8J)
- Topics
- Health Sciences › Medicine › Surgery › Surgical Simulation and Training
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Human Pose and Action Recognition
- Keywords
- surgical workflow recognition, surgical video analysis, recurrent convolutional networks, computer-assisted surgery, spatio-temporal features, Cholec80 dataset, ResNet, LSTM
- MeSH terms
- Algorithms, Humans, Image Processing, Computer-Assisted, Software, Databases, Factual, Neural Networks, Computer, Video-Assisted Surgery, Workflow
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.