List of Papers Browse by Subject Areas Author List
Abstract
Colorectal cancer screening critically depends on colonoscopy, yet existing platforms offer limited support for systematically studying the coupled dynamics of operator control, instrument motion, and visual feedback. This gap restricts reproducible closed-loop research in robotic colonoscopy, medical imaging, and emerging vision-language-action (VLA) learning paradigms. To address this challenge, we present OpenRC, an open-source modular robotic colonoscopy framework that retrofits conventional scopes while preserving clinical workflow. The framework supports simultaneous recording of video, operator commands, actuation state, and distal tip pose. We experimentally validated motion consistency and quantify cross-modal latency across sensing streams. Using this platform, we collected a multimodal dataset comprising 1,894 teleoperated episodes (~19 hours) across 10 structured task variations of routine navigation, failure events, and recovery behaviors. By unifying open hardware and an aligned multimodal dataset, OpenRC provides a reproducible foundation for research in multimodal robotic colonoscopy and surgical autonomy. Both the hardware design and dataset are available at https://github.com/artslab2019/openrc-robotic-colonoscopy. The dataset has been contributed to the Open-H-Embodiment Initiative by NVIDIA.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026/paper/5978_paper.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: https://papers.miccai.org/miccai-2026/supp/5978_supp.zip
Link to the Code Repository
https://github.com/artslab2019/openrc-robotic-colonoscopy
Link to the Dataset(s)
Robotic Colonoscopy Dataset: https://huggingface.co/datasets/nvidia/PhysicalAI-Robotics-Open-H-Embodiment/tree/main/Endoscopy/ut_austin/arts_lab/colonoscope_lerobot
BibTex
@InProceedings{KapSid_OpenRC_MICCAI2026,
author = { Kapuria, Siddhartha AND Javazm, Mohammad Rafiee AND Ikoma, Naruhiko AND Ivatury, Joga AND Nasseri, Mohammad Ali AND Navab, Nassir AND Alambeigi, Farshid},
title = { { OpenRC: An Open-Source Robotic Colonoscopy Framework for Multimodal Data Acquisition and Autonomy Research } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 16893},
month = {September},
page = {pending}
}
Reviews
Review #1
- Please describe the contribution of the paper
The paper presents an open-source robotic colonoscopy research framework designed for reproducible closed-loop experimentation and unified multimodal data collection. The system successfully synchronizes operator commands, actuation state, electromagnetically tracked distal tip pose, and video feeds with minimal latency. The utility and performance of the framework are demonstrated through assessments conducted in colon phantoms.
- Please list the major strengths of the paper: you should highlight a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
1.The paper provides a thorough and well-structured literature review, particularly regarding existing datasets in the field. 2.The work addresses a significant void in the community, as there is currently no unified, open-source hardware and software platform that enables reproducible closed-loop experimentation in colonoscopy. 3.The platform is designed to be low-cost, which increases its accessibility for research, and highly versatile, supporting the integration of several commercial colonoscopes and utilizing a modular actuation system. 4.A key technical advantage is that all data streams are timestamped using a shared system clock. Achieving precise synchronization is a highly significant and difficult task in multimodal applications, and this framework provides a robust solution to that challenge.
- Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.
1.The authors state the system aims “to balance reliable advancement with safe force application”, yet no force quantification is provided to support this. To ensure safety, the shape of the endoscope and its contact points with the colon walls should be tracked, and the actual forces should be measured to ensure they remain within a safe physiological range. 2.The feeding module relies on friction for endoscope insertion and retraction. However, the authors do not report any friction coefficients or explain how they ensure that no slippage occurs during various manipulations. 3.It would be beneficial to report the material properties of the custom silicone phantom and provide a comparison to real tissue to justify its use as a valid model. 4.Figs. 3b and c appear much earlier in the text than they are discussed, which disrupts the flow of the paper. I suggest placing them after Figure 5.In addition, the term NDI in Fig. 3 is not defined. 5.Phantoms are not ex vivo setups, they are in vitro. 6.In the segment regarding “a low-frequency sinusoidal command (0.2 Hz) along the X axis (see Fig. 2(b)) for bending”, the reference to Fig. 2(b) appears to be incorrect.
- Please rate the clarity and organization of this paper
Good
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The authors claimed to release the source code and/or dataset upon acceptance of the submission.
- Based on your review and your understanding of the MICCAI Scientific Code of Ethics, do you believe this submission may involve a potential ethics concern or violation?
N/A
- Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2026/en/REVIEWER-GUIDELINES.html
N/A
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making.
(5) Accept — should be accepted, independent of rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
The paper contributes a unified, open-source framework that addresses a significant gap in robotic colonoscopy research. The ability to synchronize multimodal data (operator commands, actuation states, and electromagnetic tracking, and video streaming) with minimal latency is a difficult technical challenge that the authors have successfully resolved. Providing a low-cost, modular platform that can be integrated with commercial colonoscopes is highly valuable for the community and will facilitate reproducible closed-loop experimentation. With the aid of computer vision, this framework has the potential to become a very important tool for developing advanced autonomous or assistive features in routine colonoscopy. While I have noted some areas for clarification regarding force quantification, friction, and overall safety, as well as proper figure placement, the overall utility and the open-source nature of the work represent a substantial contribution that justifies acceptance.
- Reviewer confidence
Very confident (4)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
Accept
- [Post rebuttal] Please justify your final decision from above.
The authors have sucessfully addressed my comments and will make the minor corrections requested.
Review #2
- Please describe the contribution of the paper
The paper presents an open-source modular robotic colonoscopy system which includes mechanical control of scope retrofits, simultaneous recording of video, operator commands, actuator state and scope distal tip pose. Main contributions include this low-cost hardware system, with motion consistency validation and cross-modal latency measurement; and a multimodal dataset collected by the system which include ~19hours recording of 10 structured tasks conducted on phantoms.
- Please list the major strengths of the paper: you should highlight a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
- Figures are well defined with clear illustration of information: Figure 1 and Figure 2 on system components; Figure 3, 4 on dataset stats and examples.
- Robotic colonoscopy with low cost and reliability/accuracy is a solid need.
- The multimodal dataset has reasonable amount of data collected and with cross-modal data collected could be a good resource for this field.
- Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.
- Dataset is all collected from phantoms and from structured tasks, which makes a large gap to real clinical translation and limits the usefulness of the dataset.
- The contribution of this hardware system and control software, may lack novelty in algorithmic aspect or methodological aspect. It may be more of an incremental work with all existing components.
- Please rate the clarity and organization of this paper
Good
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The authors claimed to release the source code and/or dataset upon acceptance of the submission.
- Based on your review and your understanding of the MICCAI Scientific Code of Ethics, do you believe this submission may involve a potential ethics concern or violation?
N/A
- Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2026/en/REVIEWER-GUIDELINES.html
N/A
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making.
(3) Weak Reject — marginally below the acceptance threshold, but would not mind if accepted, dependent on rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
Limitation of the dataset on underrepresenting real clinical scenario; Main contribution on hardware system which is integration of existing components and the dataset which lacks algorithmic/methodological novelty.
- Reviewer confidence
Very confident (4)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
N/A
- [Post rebuttal] Please justify your final decision from above.
N/A
Review #3
- Please describe the contribution of the paper
The paper presents OpenRC, an open-source framework for robotic colonoscopy research that can synchronously collect video, operator commands, actuation state, and the 6DoF pose of the distal tip. Based on this platform, the authors further developed a multimodal dataset that provides a data foundation for colonoscopy-related research.
- Please list the major strengths of the paper: you should highlight a novel formulation, an original way to use data, demonstration of clinical feasibility, a novel application, a particularly strong evaluation, or anything else that is a strong aspect of this work. Please provide details, for instance, if a method is novel, explain what aspect is novel and why this is interesting.
1.OpenRC is an open-source closed-loop platform that jointly covers observation, action, state, and pose, and it helps address a gap in the colonoscopy domain to some extent. 2.The overall system design of OpenRC is relatively well developed, covering the hardware setup, software system, and data format, with a reasonable level of completeness. 3.The constructed dataset contains diverse types of information and may offer certain value for research in this area.
- Please list the major weaknesses of the paper. Please provide details: for instance, if you state that a formulation, way of using data, demonstration of clinical feasibility, or application is not novel, then you must provide specific references to prior work.
1.The paper reads more like a description of the platform and dataset, with relatively limited system-level innovation and methodological contribution. 2.The current validation mainly focuses on motion consistency and latency, and there is still a lack of more direct evidence showing that the platform can effectively support downstream research. For example, it would be helpful to train a baseline policy or VLA on the collected data and report the corresponding performance. 3.The dataset is still limited in scale and collection diversity. If the intended use is to support large-scale VLA training, its value may be relatively limited.
- Please rate the clarity and organization of this paper
Satisfactory
- Please comment on the reproducibility of the paper. Please be aware that providing code and data is a plus, but not a requirement for acceptance.
The authors claimed to release the source code and/or dataset upon acceptance of the submission.
- Based on your review and your understanding of the MICCAI Scientific Code of Ethics, do you believe this submission may involve a potential ethics concern or violation?
N/A
- Optional: If you have any additional comments to share with the authors, please provide them here. Please also refer to our Reviewer’s guide on what makes a good review and pay specific attention to the different assessment criteria for the different paper categories: https://conferences.miccai.org/2026/en/REVIEWER-GUIDELINES.html
N/A
- Rate the paper on a scale of 1-6, 6 being the strongest (6-4: accept; 3-1: reject). Please use the entire range of the distribution. Spreading the score helps create a distribution for decision-making.
(3) Weak Reject — marginally below the acceptance threshold, but would not mind if accepted, dependent on rebuttal
- Please justify your recommendation. What were the major factors that led you to your overall score for this paper?
As noted in the weaknesses, the paper has some issues in system and methodological novelty, the depth and completeness of validation, and the value of the dataset.
- Reviewer confidence
Confident but not absolutely certain (3)
- [Post rebuttal] After reading the authors’ rebuttal, please state your final opinion of the paper.
Reject
- [Post rebuttal] Please justify your final decision from above.
I appreciate the author’s reply, but it does not resolve the concerns I raised initially. Therefore, I am maintaining my original rating.
Author Feedback
We thank the reviewers for their constructive feedback and for recognizing that OpenRC fills a gap in unified open-source platforms for reproducible closed-loop colonoscopy [R1, R2, R3], and that our dataset offers a valuable resource for the field [R1, R2, R3].
Major concerns:
Novelty and Contribution [R2.2, R3.1]: R2 and R3 viewed the work primarily as an integration of existing components with limited methodological novelty. However, to the best of our knowledge, no similar open-source, low-cost, and extensible platform exists for reproducible closed-loop colonoscopy research using existing commercial colonoscopes. The novelty of OpenRC lies in the design of the integration and actuation framework and the overall workflow specifically developed to interface with existing clinical systems rather than rebuilding an entirely new robotic platform from scratch, which we believe has not previously been demonstrated in an open-source and reproducible form for closed-loop colonoscopy research. Unlike existing systems that are often expensive and proprietary, OpenRC integrates with commercial colonoscopes for under $5,000 while preserving the clinical workflow. Besides contributing a unique multimodal dataset, another major contribution is the unified hardware-software ecosystem enabling synchronized acquisition of commands, pose, proprioception, and video streams for future research in robotics, autonomy, and AI-assisted endoscopy. We will update the text to better reflect this.
Phantom Realism [R1.3, R2.1]: R1 and R2 raised concerns regarding the phantom-only setting. As clarified in Sec. 2.4, the phantom setup was selected to enable precise 6-DoF ground-truth acquisition via EM tracking, consistent with prior colonoscopy datasets (e.g. C3VDv2 [11], SimCol3D [22]). Importantly, the phantom is fully decoupled from the proposed platform and serves as an example to demonstrate multimodal data acquisition. Our modular hardware-software framework can operate with different phantoms, colonoscopes, and image processors without modifying the overall system, enabling the community to integrate more anatomically realistic environments, sensors, and tasks as needed. Our custom phantom follows reported human colon dimensions and Paris-classified polyp morphology using silicone casting and modular 3D-printed molds, while the commercial phantom was included to increase appearance diversity. We will clarify this better in the text.
Dataset Scale and Validation Depth [R3.2, R3.3]: R3 raised concerns regarding dataset scale and the absence of trained baseline policies. While modest in scale (~19 hours, 1894 episodes), the dataset is comparable to existing colonoscopy resources [11,22], uniquely providing synchronized action-state-pose-video streams. More importantly, OpenRC is designed as a community-driven open-source platform, enabling other groups to collect compatible multimodal data using their own scopes and phantoms and continuously expand the dataset over time. We agree that training VLAs/policies is an important future research direction (Sec. 5). However, this first requires a unified, reproducible, and accessible platform that the community can broadly adopt, which is the primary objective of this work. Our validation demonstrates the reliability of this foundation through repeatable actuation (Fig. 3a) and frame-level temporal coherence across all modalities (Fig. 3b-c). We will emphasize this framing in the discussion.
Safety/Friction [R1.1, R1.2]: OpenRC preserves the colonoscope’s native compliance through adjustable preload control (Sec. 2.1). Slippage is visible during teleoperation and corrected in real time. Comprehensive shape, contact, and force tracking is planned as future work.
Minor Concerns [R1]: (R1.4) We will revisit Fig. 3 as space permits and define NDI (Northern Digital Inc.). (R1.5) We will use “in vitro” in Sec. 5.(R1.6) Fig. 2(b) shows the frame in which the X axis is defined. We will clarify in the text.
Meta-Review
Meta-review #1
- Your recommendation
Invite for Rebuttal
- Please justify your decision. In case you deviate from the reviewers’ recommendations, explain in detail the reasons why. In case of an invitation for rebuttal, clarify which points are important to address in the rebuttal.
The paper received mixed reviews. R1 is very much in favor and values the low-cost, modular platform, while some parts need clarification. R2 and R3 comment on limited system and methodological novelty as well as the limitations of the dataset. The authors are encouraged to highlight the methodological novelties including the dataset in the rebuttal.
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Accept
- Please justify your recommendation.
The authors have convincingly justified their assumptions. The work has potential impact in creating a development framework for colonoscopy-related research.
Meta-review #2
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Accept
- Please justify your recommendation.
The initial reviewers’ concerns were primarily regarding the method novelty, noting the method as combination of existing components, and limited scale and diversity of the dataset for supporting recent research such as a VLA training. The rebuttal emphasized the key contribution as reproducible system level design for closed-loop colonoscopy, emphasizing interface with clinical systems. The dataset size, while could be larger, was noted to be comparable to existing benchmarks. Rebuttal further emphasized that the proposed framework is designed to the enable community-drive open-source platform to collect compatible multimodal data using their own scopes. During the post-rebuttal, a consensus decision was not reached. Although the approach employs existing methods, it does address a critical community need for a low-cost, open-source platform, enabling reproducible research in a field which lacks large established benchmarks. In this context, the modular design serves as a merit. Thus, AC recommends acceptance, though its contingent on authors highlighting these points in the final version.
Meta-review #3
- After you have reviewed the rebuttal and updated reviews, please provide your recommendation based on all reviews and the authors’ rebuttal.
Reject
- Please justify your recommendation.
While the authors clarify that the contribution is an “open-source, low-cost platform for reproducible closed-loop colonoscopy research”, the key concerns remain unchanged: the work still “lacks novelty in algorithmic aspect or methodological aspect” and “reads more like a description of the platform and dataset.” The rebuttal does not address the “large gap to real clinical translation” from phantom-only data or the “lack of direct evidence” for downstream usefulness. Overall, the rebuttal reframes but does not resolve the main issues. The reviewers are in general in favour of Rejection.