Benchmarking and Enhancing Surgical Phase Recognition Models for Robotic-Assisted Esophagectomy

Gino Kuiper; Jasper Bongers; Jelle Ruurda; Marcel Breeuwer; Richard van Hillegersberg; Romy van Jaarsveld; Ronald de Jong; Yasmina Al Khalil; Yiping Li

REVIEW 4 major objections 4 minor 23 references

Benchmarking and Enhancing Surgical Phase Recognition Models for Robotic-Assisted Esophagectomy

T0 review · 4 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A causal transformer with hierarchical attention outperforms four prior models on a new surgical phase recognition benchmark for robotic esophagectomy.

desk verdict A useful new RAMIE phase-recognition dataset and a plausible causal-transformer adaptation, but the AutoLaparo comparison rests on literature baselines, so the unqualified superiority claim is not yet earned. read the letter →

arxiv 2412.04039 v1 pith:IY6M4IV4 submitted 2024-12-05 cs.CV

Yiping Li , Romy van Jaarsveld , Ronald de Jong , Jasper Bongers , Gino Kuiper , Richard van Hillegersberg , Jelle Ruurda , Marcel Breeuwer

show 1 more author

Yasmina Al Khalil

This is my paper · ORCID

classification cs.CV

keywords surgicalphaserecognitionrobotic-assistedesophagectomyRAMIEcausalattentionhierarchicaltransformertemporalactionsegmentationAutoLaparovideoanalysis

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to show that surgical phase recognition for robotic-assisted minimally invasive esophagectomy (RAMIE) is not served well by models built for more sequential procedures, and that a causal transformer with hierarchical attention closes the gap. To test this, the authors assembled a 27-video RAMIE dataset of the thoracic phase, with 13 annotated phases, and benchmarked four established methods plus their own model. Their model reports the best results on most metrics, with particularly large gains on segment-level metrics: Edit Score 59.50 and F1@25 58.42, versus 13.15 and 17.79 for the strongest baseline TeCNO. The stated aim is to provide a foundation for intraoperative support and postoperative analysis in RAMIE.

What carries the argument

The central object is a causal transformer with one encoder and three decoders, each with ten blocks built from causal dilated convolutions and masked self-attention. At layer $l$, the query tensor $Q_l \in \mathbb{R}^{T_l \times d \times h_l}$ and the key tensor $K_l \in \mathbb{R}^{T_l \times d \times 2h_l}$ operate on a halved sequence length with doubled head dimension, and a causal mask forces each position to attend only to earlier frames. The mechanism is adapted from ASformer, an action-segmentation transformer, and it is coupled with a loss that combines per-frame cross-entropy with a clamped temporal smoothness penalty to reduce fragmented phase predictions.

What would settle it

Re-run SV-RCNet, TeCNO, TMRNet, and Trans-SVNet with the same training/validation/test splits and the same evaluation code on both RAMIE and AutoLaparo; if the reported Edit Score gaps shrink or reverse, the claimed superiority is an artifact of protocol mismatch.

Watch

Extended reading notes

Core claim

The central claim is that adding a causal mask to hierarchical self-attention in an encoder-decoder transformer substantially improves phase recognition on procedures with repetitive phases and non-sequential transitions. On the new RAMIE dataset, the model reaches 78.28 accuracy and an Edit Score of 59.50, far above the compared baselines; on AutoLaparo it reports 83.18 accuracy versus 78.29 for Trans-SVNet. The paper attributes the gains to causal modeling of temporal order, multi-layer decoder refinement, and a smoothing loss that suppresses over-segmentation.

Load-bearing premise

The AutoLaparo comparison assumes the baseline numbers reproduced from the original paper were computed under the same metric implementation and data split, since those baselines were not re-run with the authors' evaluation code.

Editorial extensions

If this is right

If the causal architecture is the reason for the gains, temporal causality should become a standard design choice for phase-recognition models on non-sequential surgical workflows.
The RAMIE dataset and its 13-phase annotation scheme give the community a benchmark for procedures with phase-order variability, where existing benchmarks are largely sequential.
Further improvements should target phase boundaries and short phases such as right pleural and aortopulmonary lymph node dissection, where most classification errors concentrate.
Segment-level metrics like Edit Score and F1@tau should be reported alongside frame accuracy, because frame accuracy alone did not reflect the model's large improvement.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Editorial inference: If the AutoLaparo results hold under a fully re-run protocol, the same causal design may transfer to other multi-phase laparoscopic procedures with non-canonical phase orders, such as complex hysterectomy variants.
Editorial inference: The reported sensitivity to phase boundaries suggests an extension that predicts transition points explicitly, or that weights the loss by phase risk, could push clinically useful accuracy further.
Editorial inference: Because the RAMIE labels come from a single annotator, a natural next step is a multi-annotator study to measure how much of the performance gap is annotation noise rather than model behavior.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The real contribution here is the RAMIE dataset: 27 videos, 13 thoracic phases, with a sensible split and annotations guided by an expert surgeon. That alone justifies attention. The authors also did the right thing by running the four baselines from original code and evaluating everything with the Funke et al. metrics codebase, so the RAMIE head-to-head is internally controlled. The proposed model, a causal-masked variant of ASFormer, is a modest but transparent modification, and the gains on the RAMIE test set are large and consistent for edit score and F1: 59.50 vs. the best baseline's 15.63 on edit, 58.42 vs. 19.34 on F1@25, with non-overlapping standard deviations. I believe those numbers reflect a real improvement on this dataset, not just metric noise. Accuracy is essentially tied with TeCNO, which is fine—the authors are targeting temporal coherence.

The soft spots are real but concentrated. The AutoLaparo comparison in Table 3 is the load-bearing weakness: baseline numbers are taken from the original paper rather than re-run under the same Funke-code protocol, and the stress-test note is right that metric definitions, splits, and preprocessing can shift scores by several points. The authors' own standard deviation on accuracy is ±9.75, so the 4.9-point gap over Trans-SVNet is not clearly significant. Also missing: ablation of the causal mask, statistical significance tests, and a discussion of how the smooth-loss weight was chosen. The RAMIE ground truth comes from a single annotator with nine test videos, which is a limitation but not fatal for a pilot benchmark. The dataset and code are not yet public, which makes independent verification harder.

Despite the AutoLaparo issue, the paper is honest about its scope, the discussion acknowledges over-segmentation and phase-transition errors, and the methodology is described in enough detail to reproduce once the data ships. This deserves a serious referee: a good reviewer can push for the AutoLaparo rerun, ablations, and significance testing. For my own work, I would cite the RAMIE dataset once it is available; the model's novelty is not the selling point. I'd bring it to a reading group to discuss metric harmonization across surgical-phase papers, which is exactly the kind of practical issue the field needs to face.

Referee Report

4 major / 4 minor

Summary. The paper introduces a new surgical phase recognition dataset for the thoracic phase of robotic-assisted minimally invasive esophagectomy (RAMIE), comprising 27 videos with 13 annotated phases and a 14/4/9 train/validation/test split. Using this dataset, the authors benchmark four existing surgical phase recognition models (SV-RCNet, TMRNet, TeCNO, Trans-SVNet) and propose a new temporal model: an encoder-decoder transformer with causal hierarchical attention, adapted from ASFormer, trained with a cross-entropy plus smoothing loss. On RAMIE, the proposed model achieves large improvements in Edit score and F1@25/50/75 over the baselines, but its accuracy is comparable to or slightly lower than TeCNO. The paper also reports results on the public AutoLaparo dataset, where the proposed model's metrics exceed numbers taken from the original AutoLaparo paper. The authors discuss remaining challenges such as over-segmentation near phase transitions, class imbalance, and limitations of current evaluation metrics.

Significance. If the empirical results hold, the paper makes a useful contribution by providing a new dataset for a surgical procedure with complex, non-sequential phase transitions, and by showing that a causal transformer variant can substantially improve temporal segmentation metrics on that dataset. Strengths include the use of the Funke et al. evaluation codebase for all RAMIE experiments, the comparison of several baseline models run from original author code, the inclusion of Edit score and segment-level F1 metrics in addition to frame-level accuracy, and the qualitative analysis of error patterns. However, the significance is currently limited by the lack of ablations supporting the attribution of gains to the causal mechanism, by an uncontrolled AutoLaparo baseline comparison that relies on literature numbers, and by the small, single-annotator RAMIE test set. These issues are fixable and do not undermine the dataset contribution, but they must be addressed before the central claim of superiority can be accepted.

major comments (4)

[§3.5, Table 3] The AutoLaparo superiority claim is not supported by controlled evidence. Table 3 states that baseline results are taken from [12] rather than re-run under the evaluation protocol used for the proposed model (Funke et al. codebase, §3.4). Because metric definitions, train/validation splits, preprocessing, and feature extractors differ across surgical-phase papers, the baseline numbers in [12] are not directly comparable to the proposed model's scores. Moreover, §3.1.2 says the AutoLaparo split is 10/4/9 videos, which sums to 23 rather than the stated 21, and gives the same frame count (12,056) for validation and test; this internal inconsistency must be resolved. Re-running SV-RCNet, TMRNet, TeCNO, and Trans-SVNet under the paper's protocol, or at minimum reporting the exact split and metric implementations with baseline numbers, is necessary before claiming superiority on this dataset.
[§4; §3.3.2] The Discussion attributes the observed gains to the causal hierarchical attention mechanism and the smoothing loss, but no ablation isolates these components. The paper does not compare the proposed causal encoder-decoder against its non-causal ASFormer counterpart, nor against variants with λ=0 in Eq. (1), nor against configurations with different numbers of decoder layers. Without such ablations, the central claim that the causal design is responsible for the large Edit/F1 improvements on RAMIE is an unsupported attribution. Adding at least a causal-vs-non-causal comparison and a smoothing-loss on/off comparison would strengthen the paper's main conclusion.
[§3.1.1; Tables 1 and 2] The RAMIE benchmark has limited statistical grounding. The ground truth was produced by a single annotator (a PhD student guided by a medical PhD student and surgeon), with no inter-annotator agreement reported, which is important for a dataset intended as a community benchmark. The test set contains only 9 videos, and several metrics in Table 1 overlap within one standard deviation (e.g., Accuracy: Ours 78.28 ± 4.42 vs. TeCNO 78.46 ± 3.97; Jaccard: Ours 61.94 ± 7.24 vs. TeCNO 58.34 ± 4.75). The paper reports no per-video significance tests. The large Edit/F1 margins in Table 2 are encouraging, but the broad claim of superiority should be restricted to metrics where the gap is clear, or supported by significance testing.
[Abstract; §3.4] The abstract's unqualified statement that the proposed model 'demonstrates superior performance compared to existing models' is not consistent with all reported metrics. On the RAMIE dataset, the proposed model's Accuracy (78.28) is numerically lower than TeCNO's (78.46) in Table 1. The claim should be qualified to temporal segmentation metrics (Edit score, F1@τ), or the accuracy result should be discussed explicitly; otherwise the headline claim overreaches the evidence.

minor comments (4)

[§3.3.2, Eq. (1)-(2)] The loss notation is confusing: Eq. (1) uses H(S(p_{y_t,t}), y_t), but cross-entropy is written as if between a probability vector and a label without a one-hot encoding, and Eq. (2) defines Δ_t per class c while Eq. (1) applies clamp(Δ_t², 0, 16) without a class subscript. Please clarify the dimensions and define the softmax subscript.
[§3.2, §3.4] The text says baselines were implemented from original author code 'maintaining all original settings', but §3.4 says all models were evaluated with the Funke et al. codebase. Please clarify how the original settings interact with the Funke evaluation harness, especially regarding input preprocessing and feature extraction.
[Table 3] The baseline rows in Table 3 report only means, while the proposed model reports mean ± standard deviation. If the baseline numbers must remain from [12], at least state this explicitly in the text and note that standard deviations are unavailable for the baselines.
[Throughout] There are minor consistency issues in notation and naming: 'ASformer' and 'ASFormer' are used interchangeably, 'Trans-SVnet' appears in Tables 1-2 while 'Trans-SVNet' is used elsewhere, and Figure 3's l, T_l, h_l notation is defined only in the caption. A consistent notation pass would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's empirical claims rest on external baselines, a new dataset, and a transparently cited architecture adaptation.

full rationale

The manuscript makes an empirical claim: a causal hierarchical attention encoder-decoder outperforms four external baselines on a new RAMIE dataset and on the public AutoLaparo dataset. The derivation chain is not definitional. The proposed model is a neural architecture adapted transparently from ASFormer [17], an external prior work with no author overlap; the loss (cross-entropy plus a smoothness penalty) is standard and its hyperparameter is fixed before evaluation. No fitted parameter is renamed as a prediction, and no equation defines the reported metric in terms of the model's own training objective. On the RAMIE dataset, the comparison is plausibly controlled: Section 3.2 states the baselines were implemented from original author code, and Section 3.4 states all models were evaluated with the Funke et al. codebase, so the large Edit/F1 gaps in Tables 1 and 2 are internally meaningful. The AutoLaparo comparison in Table 3 does rely on baseline numbers taken from [12] rather than re-runs under the paper's own protocol, which is a legitimate validity concern about comparability of metric implementations, splits, and preprocessing; however, that is an external-baseline provenance issue, not circularity. There are no load-bearing self-citations, no uniqueness theorem imported from the authors, and no renamed known result. The central claim therefore has independent empirical content, and no circular step can be exhibited from the paper's text.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities, mediators, or conserved quantities. The only fitted free parameter is the smooth-loss weight lambda=0.15. The key assumptions are about annotation reliability, protocol-matched baselines, and the sufficiency of the two-stage training pipeline.

free parameters (1)

smooth loss weight lambda = 0.15
The hyperparameter lambda in Equation (1) is described as 'empirically set to 0.15 for all experiments' (Section 3.3.2). This is a hand-chosen value that influences the balance between classification and smoothing loss, and thus affects all reported results.

assumptions (4)

domain assumption Ground truth phase annotations are accurate and consistent
Annotations were made by a single PhD student at 25 fps, guided by a medical PhD and an expert surgeon (Section 3.1.1). No inter-annotator agreement is reported, so annotation noise is unquantified.
domain assumption The AutoLaparo baseline results from [12] are directly comparable to the authors' own evaluation
Table 3 cites [12] for baseline means, while the text claims to use the Funke et al. [21] evaluation codebase. If training/validation splits or metric implementations differ, the comparison is invalid.
domain assumption Two-stage training with frozen ResNet features is sufficient
The temporal model consumes pre-extracted ResNet50 features (Section 3.3.3), so errors in the feature extractor propagate. The paper does not explore end-to-end training.
ad hoc to paper The causal transformer is a faithful adaptation of ASFormer's hierarchical attention
The paper states it 'implemented a causal transformer architecture' inspired by ASFormer, but no code or ablation verifies that other design choices (e.g., number of decoder layers, dilated convolution settings) match ASFormer.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Benchmarking and Enhancing Surgical Phase Recognition Models for Robotic-Assisted Esophagectomy." pith.science (2026). https://pith.science/paper/IY6M4IV4

@misc{pith2026241204039,
  author       = {Pith},
  title        = {Pith review of: Benchmarking and Enhancing Surgical Phase Recognition Models for Robotic-Assisted Esophagectomy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IY6M4IV4}},
  note         = {Machine review of arXiv:2412.04039}
}

read the original abstract

Robotic-assisted minimally invasive esophagectomy (RAMIE) is a recognized treatment for esophageal cancer, offering better patient outcomes compared to open surgery and traditional minimally invasive surgery. RAMIE is highly complex, spanning multiple anatomical areas and involving repetitive phases and non-sequential phase transitions. Our goal is to leverage deep learning for surgical phase recognition in RAMIE to provide intraoperative support to surgeons. To achieve this, we have developed a new surgical phase recognition dataset comprising 27 videos. Using this dataset, we conducted a comparative analysis of state-of-the-art surgical phase recognition models. To more effectively capture the temporal dynamics of this complex procedure, we developed a novel deep learning model featuring an encoder-decoder structure with causal hierarchical attention, which demonstrates superior performance compared to existing models.

Figures

Figures reproduced from arXiv: 2412.04039 by the authors.

**Figure 2.** Number of frames per phase in RAMIE dataset [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Proposed model architecture (left) and comparison of hierarchical attention in causal and non-causal [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Mean F1 scores across surgical phases in RAMIE dataset [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

**Figure 5.** Figure 5: Qualitative Result on RAMIE dataset [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

23 extracted references · 15 canonical work pages

[12]

Autolaparo: A new dataset of integrated multi-tasks for image-guided surgical automation in laparoscopic hysterectomy

Ziyi Wang, Bo Lu, Yonghao Long, Fangxun Zhong, Tak-Hong Cheung, Qi Dou, and Yunhui Liu. Autolaparo: A new dataset of integrated multi-tasks for image-guided surgical automation in laparoscopic hysterectomy. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pages 486–

work page
[1]

Global cancer statistics 2022: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries

Freddie Bray, Mathieu Laversanne, Hyuna Sung, Jacques Ferlay, Rebecca L Siegel, Isabelle Soerjomataram, and Ahmedin Jemal. Global cancer statistics 2022: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians, 74(3):229–263, 2024

2022
[2]

Robotic-assisted mini- mally invasive esophagectomy (ramie) for esophageal cancer training curriculum—a worldwide delphi con- sensus study

Hans F Fuchs, Justin W Collins, Benjamin Babic, Christopher DuCoin, Ozanan R Meireles, Peter P Grim- minger, Matthew Read, Abbas Abbas, Rubens Sallum, Beat P M¨ uller-Stich, et al. Robotic-assisted mini- mally invasive esophagectomy (ramie) for esophageal cancer training curriculum—a worldwide delphi con- sensus study. Diseases of the Esophagus, 35(6):doa...

work page 2022
[3]

Learning curve for robot-assisted minimally invasive thoracoscopic esophagectomy: results from 312 cases

Pieter C van der Sluis, Jelle P Ruurda, Sylvia van der Horst, Lucas Goense, and Richard van Hillegersberg. Learning curve for robot-assisted minimally invasive thoracoscopic esophagectomy: results from 312 cases. The Annals of Thoracic Surgery, 106(1):264–271, 2018

work page 2018
[4]

Deep learning-based recognition of key anatomical structures during robot-assisted minimally invasive esophagectomy

RB Den Boer, TJM Jaspers, C De Jongh, JPW Pluim, F Van Der Sommen, T Boers, R van Hillegersberg, MAJM Van Eijnatten, and JP Ruurda. Deep learning-based recognition of key anatomical structures during robot-assisted minimally invasive esophagectomy. Surgical endoscopy, 37(7):5164–5175, 2023

work page 2023
[5]

Real-time detection of the recurrent laryngeal nerve in thoracoscopic esophagectomy using artificial intelligence

Kazuma Sato, Takeo Fujita, Hiroki Matsuzaki, Nobuyoshi Takeshita, Hisashi Fujiwara, Shuichi Mitsunaga, Takashi Kojima, Kensaku Mori, and Hiroyuki Daiko. Real-time detection of the recurrent laryngeal nerve in thoracoscopic esophagectomy using artificial intelligence. Surgical Endoscopy, 36(7):5531–5539, 2022

work page 2022
[6]

Automated surgical-phase recognition for robot-assisted minimally invasive esophagectomy using artificial intelligence

Masashi Takeuchi, Hirofumi Kawakubo, Kosuke Saito, Yusuke Maeda, Satoru Matsuda, Kazumasa Fukuda, Rieko Nakamura, and Yuko Kitagawa. Automated surgical-phase recognition for robot-assisted minimally invasive esophagectomy using artificial intelligence. Annals of Surgical Oncology, 29(11):6847–6855, 2022

work page 2022
[7]

Tecno: Surgical phase recognition with multi-stage temporal convolutional net- works

Tobias Czempiel, Magdalini Paschali, Matthias Keicher, Walter Simson, Hubertus Feussner, Seong Tae Kim, and Nassir Navab. Tecno: Surgical phase recognition with multi-stage temporal convolutional net- works. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings,...

work page 2020

Show all 23 references

[8]

Active learning for extracting surgomic features in robot-assisted minimally invasive esophagectomy: a prospective annotation study

Johanna M Brandenburg, Alexander C Jenke, Antonia Stern, Marie TJ Daum, Andr´ e Schulze, Rayan Younis, Philipp Petrynowski, Tornike Davitashvili, Vincent Vanat, Nithya Bhasker, et al. Active learning for extracting surgomic features in robot-assisted minimally invasive esophag...

2023
[9]

Surgical data science-from concepts to clinical translation

Lena Maier-Hein, Matthias Eisenmann, Duygu Sarikaya, Keno M¨ arz, Toby Collins, Anand Malpani, Jo- hannes Fallert, Hubertus Feussner, Stamatia Giannarou, Pietro Mascagni, et al. Surgical data science-from concepts to clinical translation. arXiv preprint arXiv:2011.02284, 2, 2020

2011 arXiv
[10]

Outcomes and complications after robot- assisted minimally invasive esophagectomy

Gijsbert van Boxel, Richard van Hillegersberg, and Jelle Ruurda. Outcomes and complications after robot- assisted minimally invasive esophagectomy. Journal of Visualized Surgery, 5, 2019

2019
[11]

A standardized approach for the thoracic dissection in robotic-assisted minimally invasive esophagectomy (ramie).Diseases of the Esophagus, 33(Supplement 2):doaa066, 2020

BF Kingma, M Read, R Van Hillegersberg, YK Chao, and JP Ruurda. A standardized approach for the thoracic dissection in robotic-assisted minimally invasive esophagectomy (ramie).Diseases of the Esophagus, 33(Supplement 2):doaa066, 2020

2020
[13]

Sv-rcnet: workflow recognition from surgical videos using recurrent convolutional network

Yueming Jin, Qi Dou, Hao Chen, Lequan Yu, Jing Qin, Chi-Wing Fu, and Pheng-Ann Heng. Sv-rcnet: workflow recognition from surgical videos using recurrent convolutional network. IEEE transactions on medical imaging, 37(5):1114–1126, 2017

2017
[14]

Temporal memory relation network for workflow recognition from surgical video

Yueming Jin, Yonghao Long, Cheng Chen, Zixu Zhao, Qi Dou, and Pheng-Ann Heng. Temporal memory relation network for workflow recognition from surgical video. IEEE Transactions on Medical Imaging, 40 (7):1911–1923, 2021

1911
[15]

Trans-svnet: Accurate phase recognition from surgical videos via hybrid embedding aggregation transformer

Xiaojie Gao, Yueming Jin, Yonghao Long, Qi Dou, and Pheng-Ann Heng. Trans-svnet: Accurate phase recognition from surgical videos via hybrid embedding aggregation transformer. InMedical Image Computing and Computer Assisted Intervention–MICCAI 2021: 24th International Conferenc...

2021
[16]

Endonet: a deep architecture for recognition tasks on laparoscopic videos

Andru P Twinanda, Sherif Shehata, Didier Mutter, Jacques Marescaux, Michel De Mathelin, and Nicolas Padoy. Endonet: a deep architecture for recognition tasks on laparoscopic videos. IEEE transactions on medical imaging, 36(1):86–97, 2016

2016
[17]

Asformer: Transformer for action segmentation

Fangqiu Yi, Hongyu Wen, and Tingting Jiang. Asformer: Transformer for action segmentation. arXiv preprint arXiv:2110.08568, 2021

2021 arXiv
[18]

Kuehne, A

H. Kuehne, A. B. Arslan, and T. Serre. The language of actions: Recovering the syntax and semantics of goal-directed human activities. In Proceedings of Computer Vision and Pattern Recognition Conference (CVPR), 2014

2014
[19]

Combining embedded accelerometers with computer vision for recognizing food preparation activities

Sebastian Stein and Stephen J McKenna. Combining embedded accelerometers with computer vision for recognizing food preparation activities. In Proceedings of the 2013 ACM international joint conference on Pervasive and ubiquitous computing, pages 729–738, 2013

2013
[20]

Ms-tcn: Multi-stage temporal convolutional network for action seg- mentation

Yazan Abu Farha and Jurgen Gall. Ms-tcn: Multi-stage temporal convolutional network for action seg- mentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 3575–3584, 2019

2019
[21]

Metrics matter in surgical phase recognition

Isabel Funke, Dominik Rivoir, and Stefanie Speidel. Metrics matter in surgical phase recognition. arXiv preprint arXiv:2305.13961, 2023

2023 arXiv
[22]

Segmental spatiotemporal cnns for fine- grained action segmentation

Colin Lea, Austin Reiter, Ren´ e Vidal, and Gregory D Hager. Segmental spatiotemporal cnns for fine- grained action segmentation. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part III 14, pages 36–52. Spr...

2016
[23]

Temporal convolutional networks for action segmentation and detection

Colin Lea, Michael D Flynn, Rene Vidal, Austin Reiter, and Gregory D Hager. Temporal convolutional networks for action segmentation and detection. In proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 156–165, 2017

2017

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [12]

Autolaparo: A new dataset of integrated multi-tasks for image-guided surgical automation in laparoscopic hysterectomy

Ziyi Wang, Bo Lu, Yonghao Long, Fangxun Zhong, Tak-Hong Cheung, Qi Dou, and Yunhui Liu. Autolaparo: A new dataset of integrated multi-tasks for image-guided surgical automation in laparoscopic hysterectomy. In International Conference on Medical Image Computing and Computer-Assisted Intervention, pages 486–

work page

[2] [1]

Global cancer statistics 2022: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries

Freddie Bray, Mathieu Laversanne, Hyuna Sung, Jacques Ferlay, Rebecca L Siegel, Isabelle Soerjomataram, and Ahmedin Jemal. Global cancer statistics 2022: Globocan estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: a cancer journal for clinicians, 74(3):229–263, 2024

2022

[3] [2]

Robotic-assisted mini- mally invasive esophagectomy (ramie) for esophageal cancer training curriculum—a worldwide delphi con- sensus study

Hans F Fuchs, Justin W Collins, Benjamin Babic, Christopher DuCoin, Ozanan R Meireles, Peter P Grim- minger, Matthew Read, Abbas Abbas, Rubens Sallum, Beat P M¨ uller-Stich, et al. Robotic-assisted mini- mally invasive esophagectomy (ramie) for esophageal cancer training curriculum—a worldwide delphi con- sensus study. Diseases of the Esophagus, 35(6):doa...

work page 2022

[4] [3]

Learning curve for robot-assisted minimally invasive thoracoscopic esophagectomy: results from 312 cases

Pieter C van der Sluis, Jelle P Ruurda, Sylvia van der Horst, Lucas Goense, and Richard van Hillegersberg. Learning curve for robot-assisted minimally invasive thoracoscopic esophagectomy: results from 312 cases. The Annals of Thoracic Surgery, 106(1):264–271, 2018

work page 2018

[5] [4]

Deep learning-based recognition of key anatomical structures during robot-assisted minimally invasive esophagectomy

RB Den Boer, TJM Jaspers, C De Jongh, JPW Pluim, F Van Der Sommen, T Boers, R van Hillegersberg, MAJM Van Eijnatten, and JP Ruurda. Deep learning-based recognition of key anatomical structures during robot-assisted minimally invasive esophagectomy. Surgical endoscopy, 37(7):5164–5175, 2023

work page 2023

[6] [5]

Real-time detection of the recurrent laryngeal nerve in thoracoscopic esophagectomy using artificial intelligence

Kazuma Sato, Takeo Fujita, Hiroki Matsuzaki, Nobuyoshi Takeshita, Hisashi Fujiwara, Shuichi Mitsunaga, Takashi Kojima, Kensaku Mori, and Hiroyuki Daiko. Real-time detection of the recurrent laryngeal nerve in thoracoscopic esophagectomy using artificial intelligence. Surgical Endoscopy, 36(7):5531–5539, 2022

work page 2022

[7] [6]

Automated surgical-phase recognition for robot-assisted minimally invasive esophagectomy using artificial intelligence

Masashi Takeuchi, Hirofumi Kawakubo, Kosuke Saito, Yusuke Maeda, Satoru Matsuda, Kazumasa Fukuda, Rieko Nakamura, and Yuko Kitagawa. Automated surgical-phase recognition for robot-assisted minimally invasive esophagectomy using artificial intelligence. Annals of Surgical Oncology, 29(11):6847–6855, 2022

work page 2022

[8] [7]

Tecno: Surgical phase recognition with multi-stage temporal convolutional net- works

Tobias Czempiel, Magdalini Paschali, Matthias Keicher, Walter Simson, Hubertus Feussner, Seong Tae Kim, and Nassir Navab. Tecno: Surgical phase recognition with multi-stage temporal convolutional net- works. In Medical Image Computing and Computer Assisted Intervention–MICCAI 2020: 23rd International Conference, Lima, Peru, October 4–8, 2020, Proceedings,...

work page 2020