@CammaLabi
iAccount based inUnited States
About this account
- Account based in
- United States
- Connected via
- United States Android App
Account-level information from X, not a live location or the device used for a specific post.
Computational Analysis and Modeling of Medical Activities research group developing #SurgicalDataScience solutions at @unistra, @IHUStrasbourg
Strasbourg, France
Joined September 2020
- Tweets275
- Following94
- Followers503
- Likes203
Pinned Tweet
CAMMA develops #ComputerVision & #MachineLearning methods and applications to perceive, model and analyse the vast amount of digital data generated in ORs in order to support surgical teams delivering safe and efficient care. #SurgicalDataScience #SurgicalAI #AI4Surgery
1/3🚨 New Annotations Released: MultiBypass140 Now Includes Intraoperative Adverse Events (iAEs) 🚨
#SurgicalAI #MedicalAI #LaparoscopicSurgery #DatasetRelease
2/3 Updates include:
🔹Types: Bleeding, Thermal Injury, Mechanical Injury, Ischemic Injury, Anastomotic Insufficiency
🌡️Severity Grades: up to 5 levels per type
✅Rectification Status: Complete or Incomplete
#SurgicalAI #MedicalAI #LaparoscopicSurgery #DatasetRelease
3/3 These annotations enable granular analysis of surgical performance and iAEs.
🔗 Dataset: github.com/CAMMA-public/Mult…
#SurgicalAI #MedicalAI #LaparoscopicSurgery #DatasetRelease
@Nwoyecid
1/3🚀#CVPR2025 Accepted + Dataset Release!
Presenting CholecTrack20—the first surgical dataset with multi-perspective tool tracking (intraoperative/intracorporeal/visibility). Dataset made public today under CC-BY-NC-SA 4.0!
🔗 bit.ly/ckt20
2/3📊Dataset Highlights:
✅20 full-length videos
✅35K+ frames, 65K+ tool instances
✅Annotations: spatial, category, ID, operator, phase, & visual challenges
✅Considers out-of-body, out-of-camera view/entry/exit of tools
#SurgicalAI #OpenScience #ComputerVision #AI4Surgery
3/3🔎Why it matters:
Current methods (<45% HOTA) fail on surgical-specific cases like tools exiting the body. Our dataset enables development of clinically-relevant tracking for:
✔️Skill assessment
✔️Safety zone estimation
✔️Human-machine collaboration
#MedTech #ComputerVision
📢 Thrilled to launch the 2025 #SAGES CVS Challenge, a #LighthouseChallenge at #MICCAI2025 in Daejeon, South Korea! 🌟 Register now at cvschallenge.org to build AI solutions for Critical View of Safety (CVS) assessment in laparoscopic cholecystectomy! 🏥🔬(1/5)
🔍 With these additions, in 2025 we are honored to be recognized as a Lighthouse Challenge (tinyurl.com/bddsv3ne)! These challenges undergo a rigorous review process, representing the gold-standard for Medical AI challenges with major clinical impact potential (4/5)
🙌 Huge thanks to our amazing organizing team! @theCVSchallenge @p_masca @dpkalapatt @aditya_murali7 @Laparoscopes @Guy_Rosman7 @Oz_Meireles @SarahChoksi @SAGES_Updates (5/5)
2/2
#SurgiTrack:
🔹 Handles occlusions, OOCV, OOB
🔹 Includes multi-perspective trajectories
🔹 Obtains 92% re-ID success
📄 Published in @MedImageAnalysis
🖊️@Nwoyecid and Nicolas Padoy
🌐 bit.ly/surgitrack
#AI #Surgery #MedTech #IPCAI
1/2
🎯 New Paper Alert! 🏥✨
#SurgiTrack is a breakthrough in surgical tool tracking! Leveraging direction-based Re-ID & hierarchical bipartite graphs, it achieves 62.8% HOTA on #CholecTrack20, outperforming SOTA methods.
#AI #Surgery #MedTech #IPCAI
🚀 Just published! Our latest research presents SAF-IS, a novel framework for instance segmentation of surgical tools not requiring spatial annotations to train! 🏥🤖
📖 Read more in Elsevier Medical Image Analysis: doi.org/10.1016/j.media.2025…
CAMMA retweeted
🚀 Excited to share our latest research @CammaLab: SurgiTrack - Fine-grained multi-class multi-tool tracking in surgical videos! 🎥🔧
✅ Fine-grained multi-perspective trajectories
✅ Direction-based re-ID for similar tools
✅ Real-time inference
Follow along to learn more!🧵1/11
🚀 The 4th Surgical Data Science Summer School is back! Join us July 7–11, 2025, at IHU Strasbourg for hands-on AI & endoscopic video analysis with top experts.
👉 Places are limited! Apply by April 1, 2025: edu4sds.org/
#SurgicalDataScience #AI
CAMMA retweeted
Can you spot #AI generated surgical images?
Surgeons please help us evaluate a novel surgical text-to-image gen model by completing a short survey 🧑⚕️👩⚕️
Survey 👉 t.ly/endogen
ArXiv 👉 arxiv.org/abs/2407.09230
#SurgicalDataScience #GenAI #medtwitter #SoMe4Surgery