CAMMA

Computational Analysis and Modeling of Medical Activities

Coordinator (person and lab): Nicolas Padoy, ICUBE

CAMI Partners : LaTIM Brest (XAware), LTSI & TIMC & LIRMM (M2CAMI workgroup), ISIR laboratory (Paris)

Started: September 2013

Project websitehttp://camma.u-strasbg.fr/


CAMMA project

Motivations and objectives:   

The Operating Room (OR) is a high-tech environment that strongly relies on electronic devices, information systems and video equipment, all of which exchange, process or display signals containing key data about the underlying surgical process. In spite of their potential utility in contributing to the development of intelligent applications for the OR, currently these signals are being used solely for the immediate performance of the surgery. The fundamental idea behind our research is the fact that automatically recording, processing and analyzing the large amount of multi-modal surgical signals available in the OR, an endeavor largely unattempted so far, can lead to new assistance and decision support tools for surgeons and surgical staff.

XAware

Consequently, the research group on Computational Modeling and Analysis of Medical Activities aims at developing a multi-sensor data acquisition system for the operating room (OR) as well as devising novel methods to perceive, model and analyze surgical activities. These methods will be used to design a new generation of context-aware assistance systems for the OR that can help the surgical team with performing routine tasks and also enable them to better process and visualize the large amount of information generated during a surgical procedure.

2
Clinical tracking

Our research lies at the intersection of computer vision, medical robotics, augmented reality and machine learning. We seek in particular to leverage the large amount of data collected from medical devices and sensors used in a wide variety of surgical procedures to develop new algorithms and solutions for activity recognition and modeling. 

1
Surgical database analysis

A first application of our research is the monitoring of radiation during image-guided interventions to address the increased exposure of surgical staff to X-rays generated by imaging devices. This work is being carried out in collaboration with Strasbourg’s University Hospital and the IHU MixSurg institute.

A second application (video) is the development of tools to automatically process endoscopic surgeries for real-time safety monitoring, video database indexing and surgical education. This work is being carried out in collaboration with Strasbourg’s University Hospital, the IRCAD and the IHU MixSurg institute.


Main results and on-going work  

All results and on going work are available on the dedicated CAMMA website  via the several presented videos.

You can know more on the current news and events of the CAMMA project here.

Our team is collaborating with the University Hospital of Strasbourg, IHU Strasbourg and IRCAD to build datasets for various medical recognition tasks which you can consult on CAMMA Website / Datasets.

 

 


Recent publications:

G. Yengera, D. Mutter, J. Marescaux, N. Padoy, Less is More: Surgical Phase Recognition with Less Annotations through Self-Supervised Pre-training of CNN-LSTM Networks, arXiv:1805.08569, 2018
A. Kadkhodamohammadi, N. Padoy, A generalizable approach for multi-view 3D human pose regression, arXiv:1804.10462, 2018
A.P. Twinanda, G. Yengera, D. Mutter, J. Marescaux, N. Padoy, RSDNet: Learning to Predict Remaining Surgery Duration from Laparoscopic Videos Without Manual Annotations, IEEE Transactions on Medical Imaging (TMI), to appear, arXiv preprint, 2018
A. Vardazaryan, D. Mutter, J. Marescaux, N. Padoy, Weakly-Supervised Learning for Tool Localization in Laparoscopic Videos, MICCAI-LABELS, arXiv preprint, 2018
V. Srivastav, T. Issenhuth, A. Kadkhodamohammadi, M. de Mathelin, A. Gangi, N. Padoy, MVOR: A Multi-view RGB-D Operating Room Dataset for 2D and 3D Human Pose Estimation, MICCAI-LABELS, arXiv preprint, 2018
N. Loy Rodas, A. Kadkhodamohammadi, M. de Mathelin, A. Gangi, N. Padoy, A Global Radiation Awareness System using Augmented Reality and Monte Carlo Simulations, European Congress of Radiology (ECR), 2018
N. Loy Rodas, P. Ghimire, M. de Mathelin, A. Gangi, N. Padoy, Teaching Radiation Safety Intuitively with a Head-Mounted Display, European Congress of Radiology (ECR), 2018

All CAMMA publications are listed in CAMMA website.


CAMMA Team:

Group Leader
Nicolas Padoy npadoy@unistra.fr
Research Engineers
Nicolas Loy Rodas nloyrodas@unistra.fr
Armine Vardazaryan vardazaryan@unistra.fr
Gaurav Yengera yengera@unistra.fr
Postdoctoral Fellows
Rahim Kadkhodamohammadi
PhD Students
Chinedu Nwoye
Vinkle Srivastav
Tong Yu
Mario Aricò (with ISIR) arico@isir.upmc.fr
3
CAMMA Team

 

 

logo partenaire