Analyse de vidéos de microscopes chirurgicaux pour la reconnaissance automatique d'étapes en combinant SVM et HMM

Florent Lalys 1 Laurent Riffaud 2 Xavier Morandi 1, 2 Pierre Jannin 1
1 VisAGeS - Vision, Action et Gestion d'informations en Santé
INSERM - Institut National de la Santé et de la Recherche Médicale : U746, Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : In image-guided surgery, the automatic extraction of information from the Operating Room (OR) has recently gained much interest. In particular, the automatic recognition of surgical phases, or more generally surgical events, allows bringing an additional help to surgeons, for intra or post-operative use. Clinical applications include the creation of post-operative reports, teaching, learning, surgical assessment, optimisation of OR management and of surgeries. In this article, we present a novel approach that focused on the automatic recognition of phases by microscope image analysis, which has never been done before. We used a hybrid method that combines supervised classification to extract binary visual cues and a discrete Hidden Markov Model to take into account the temporal aspect. Our framework was tested on two various datasets, including one specific type of neurosurgical intervention and one type of ophthalmological surgery. Cross-validation studies were carried out to find recognition rates of 90% for the first dataset and 93% for the second one. This will enable the system to be used in clinical applications such as post-operative surgical video indexation.
Document type :
Conference papers
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download

https://hal.inria.fr/inria-00595720
Contributor : Peter Sturm <>
Submitted on : Wednesday, May 25, 2011 - 2:05:56 PM
Last modification on : Tuesday, March 5, 2019 - 1:45:46 AM
Long-term archiving on : Friday, August 26, 2011 - 2:29:21 AM

File

paper19.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00595720, version 1

Citation

Florent Lalys, Laurent Riffaud, Xavier Morandi, Pierre Jannin. Analyse de vidéos de microscopes chirurgicaux pour la reconnaissance automatique d'étapes en combinant SVM et HMM. ORASIS - Congrès des jeunes chercheurs en vision par ordinateur, INRIA Grenoble Rhône-Alpes, Jun 2011, Praz-sur-Arly, France. epub ahead of print. ⟨inria-00595720⟩

Share

Metrics

Record views

1398

Files downloads

315