A polynomial model of surgical gestures for real-time retrieval of surgery videos

Authors:
Gwénolé Quellec;Mathieu Lamard;Zakarya Droueche;Béatrice Cochener;Christian Roux;Guy Cazuguel
Affiliations:
Inserm, UMR 1101, Brest, France;Inserm, UMR 1101, Brest, France, Univ Bretagne Occidentale, Brest, France;Inserm, UMR 1101, Brest, France, Dpt. ITI, INSTITUT. TELECOM, TELECOM Bretagne, UEB, Brest, France;Inserm, UMR 1101, Brest, France, Univ Bretagne Occidentale, Brest, France, CHU Brest, Service d'Ophtalmologie, Brest, France;Inserm, UMR 1101, Brest, France, Dpt. ITI, INSTITUT. TELECOM, TELECOM Bretagne, UEB, Brest, France;Inserm, UMR 1101, Brest, France, Dpt. ITI, INSTITUT. TELECOM, TELECOM Bretagne, UEB, Brest, France
Venue:
MCBR-CDS'12 Proceedings of the Third MICCAI international conference on Medical Content-Based Retrieval for Clinical Decision Support
Year:
2012

Citing 15
Cited 1

A framework for multiple-instance learning

NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
On Space-Time Interest Points

International Journal of Computer Vision
Evaluation campaigns and TRECVid

MIR '06 Proceedings of the 8th ACM international workshop on Multimedia information retrieval
Segmenting, Modeling, and Matching Video Clips Containing Multiple Moving Objects

IEEE Transactions on Pattern Analysis and Machine Intelligence
Towards optimal bag-of-features for object categorization and semantic video retrieval

Proceedings of the 6th ACM international conference on Image and video retrieval
Motion segmentation and retrieval for 3D video based on modified shape distribution

EURASIP Journal on Applied Signal Processing
Detecting repeats for video structuring

Multimedia Tools and Applications
Laparoscopic Tool Tracking Method for Augmented Reality Surgical Applications

ISBMS '08 Proceedings of the 4th international symposium on Biomedical Simulation
Video Event Recognition Using Kernel Methods with Multilevel Temporal Alignment

IEEE Transactions on Pattern Analysis and Machine Intelligence
Hidden Markov Model for Content-Based Video Retrieval

AMS '09 Proceedings of the 2009 Third Asia International Conference on Modelling & Simulation
Speeding up and boosting diverse density learning

DS'10 Proceedings of the 13th international conference on Discovery science
An image retrieval approach to setup difficulty levels in training systems for endomicroscopy diagnosis

MICCAI'10 Proceedings of the 13th international conference on Medical image computing and computer-assisted intervention: Part II
Content-based surgical workflow representation using probabilistic motion modeling

MIAR'10 Proceedings of the 5th international conference on Medical imaging and augmented reality
Real-Time endoscopic mosaicking

MICCAI'06 Proceedings of the 9th international conference on Medical Image Computing and Computer-Assisted Intervention - Volume Part I
Semantic-Based Surveillance Video Retrieval

IEEE Transactions on Image Processing

Overview of the third workshop on medical content---based retrieval for clinical decision support (MCBR---CDS 2012)

MCBR-CDS'12 Proceedings of the Third MICCAI international conference on Medical Content-Based Retrieval for Clinical Decision Support

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper introduces a novel retrieval framework for surgery videos. Given a query video, the goal is to retrieve videos in which similar surgical gestures appear. In this framework, the motion content of short video subsequences is modeled, in real-time, using spatiotemporal polynomials. The retrieval engine needs to be trained: key spatiotemporal polynomials, characterizing semantically-relevant surgical gestures, are identified through multiple-instance learning. Then, videos are compared in a high-level space spanned by these key spatiotemporal polynomials. The framework was applied to a dataset of 900 manually-delimited clips from 100 cataract surgery videos. High classification performance (Az=0.816±0.118) and retrieval performance (MAP=0.358) were observed.