March 2012
Volume 53, Issue 14
Free
ARVO Annual Meeting Abstract  |   March 2012
Computer-aided Retinal Surgery Using Video Data Compression
Author Affiliations & Notes
  • Mathieu Lamard
    LaTIM, Universite de Brest, Brest, France
  • Béatrice Cochener
    LaTIM, Universite de Brest, Brest, France
    Ophtalmology, Chu de Brest, Brest, France
  • Gwenolé Quellec
    LaTIM, Universite de Brest, Brest, France
  • Guy Cazuguel
    LaTIM, Universite de Brest, Brest, France
    telecom Bretagne, UEB, Institut telecom, Brest, France
  • Christian Roux
    LaTIM, Universite de Brest, Brest, France
    telecom Bretagne, UEB, Institut telecom, Brest, France
  • Zakarya Droueche
    LaTIM, Universite de Brest, Brest, France
  • Footnotes
    Commercial Relationships  Mathieu Lamard, None; Béatrice Cochener, None; Gwenolé Quellec, None; Guy Cazuguel, None; Christian Roux, None; Zakarya Droueche, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science March 2012, Vol.53, 3085. doi:https://doi.org/
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Mathieu Lamard, Béatrice Cochener, Gwenolé Quellec, Guy Cazuguel, Christian Roux, Zakarya Droueche; Computer-aided Retinal Surgery Using Video Data Compression. Invest. Ophthalmol. Vis. Sci. 2012;53(14):3085. doi: https://doi.org/.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: : This paper introduces ongoing research on computer-aided ophthalmic (retinal) surgery. A content-based video retrieval (CBVR) system is presented: given a video stream captured by a digital camera monitoring the current epiretinal membrane surgery, the system retrieves similar videos in video archives. These information could guide the surgery steps or even generate surgical alerts if the current surgery shares complications with already archived videos or let the surgeon know what a more experienced fellow worker would do in a similar situation (recommendation generation).

Methods: : We propose to use data compression to extract video features. First, motion vectors are derived from the standard ‘MPEG-4 AVC/H.264’ video stream. Second, motion-based image sequence segmentation is performed by a combination of k-means clustering and motion consistency verification. Third, we used the well-known Kalman filter to track region displacements between consecutive frames and therefore characterize region trajectories. Finally, we combined this motion information with residual information consisting of the difference between original input images and predicted images. To compare videos, we adopted an extension of fast dynamic time warping (EFDTW) to multidimensional time series.

Results: : The system was applied to a small dataset of 24 video-recorded retinal surgeries. Videos have an average length of (621s +- 299s) and images have a definition of 720x576 pixels. An ophthalmic surgeon has divided each video into three new videos, each corresponding to one step of the membrane peeling procedure: Injection, Coat and Vitrectomy. As a result, 72 videos have been obtained and each video is associated with one class (Injection, Coat or Vitrectomy). The effectiveness of the proposed method, measured by the area under the Receiver Operating Characteristic curve, is interesting (Az ≈ 0.73).

Conclusions: : A novel CBVR system, allowing retrieval of medical video, has been presented. Experiments on the dataset of retinal surgery steps validate the semantic relevance of retrieved results in ophthalmic applications. A larger cataract surgery dataset is currently being collected and interpreted at the LaTIM laboratory; it will allow such an experiment in future works.

Keywords: image processing • retina 
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×