Literature DB >> 28357628

Addressing multi-label imbalance problem of surgical tool detection using CNN.

Manish Sahu1, Anirban Mukhopadhyay2, Angelika Szengel2, Stefan Zachow2.   

Abstract

PURPOSE: A fully automated surgical tool detection framework is proposed for endoscopic video streams. State-of-the-art surgical tool detection methods rely on supervised one-vs-all or multi-class classification techniques, completely ignoring the co-occurrence relationship of the tools and the associated class imbalance.
METHODS: In this paper, we formulate tool detection as a multi-label classification task where tool co-occurrences are treated as separate classes. In addition, imbalance on tool co-occurrences is analyzed and stratification techniques are employed to address the imbalance during convolutional neural network (CNN) training. Moreover, temporal smoothing is introduced as an online post-processing step to enhance runtime prediction.
RESULTS: Quantitative analysis is performed on the M2CAI16 tool detection dataset to highlight the importance of stratification, temporal smoothing and the overall framework for tool detection.
CONCLUSION: The analysis on tool imbalance, backed by the empirical results, indicates the need and superiority of the proposed framework over state-of-the-art techniques.

Keywords:  CNN; Laparoscopic videos; Multi-label learning; Surgical tool detection; Transfer learning

Mesh:

Year:  2017        PMID: 28357628     DOI: 10.1007/s11548-017-1565-x

Source DB:  PubMed          Journal:  Int J Comput Assist Radiol Surg        ISSN: 1861-6410            Impact factor:   2.924


  8 in total

1.  Statistical modeling and recognition of surgical workflow.

Authors:  Nicolas Padoy; Tobias Blum; Seyed-Ahmad Ahmadi; Hubertus Feussner; Marie-Odile Berger; Nassir Navab
Journal:  Med Image Anal       Date:  2010-12-08       Impact factor: 8.545

2.  Modeling and segmentation of surgical workflow from laparoscopic video.

Authors:  Tobias Blum; Hubertus Feussner; Nassir Navab
Journal:  Med Image Comput Comput Assist Interv       Date:  2010

3.  Detecting Surgical Tools by Modelling Local Appearance and Global Shape.

Authors:  David Bouget; Rodrigo Benenson; Mohamed Omran; Laurent Riffaud; Bernt Schiele; Pierre Jannin
Journal:  IEEE Trans Med Imaging       Date:  2015-12       Impact factor: 10.048

4.  Surgical gesture classification from video and kinematic data.

Authors:  Luca Zappella; Benjamín Béjar; Gregory Hager; René Vidal
Journal:  Med Image Anal       Date:  2013-04-28       Impact factor: 8.545

5.  Fast part-based classification for instrument detection in minimally invasive surgery.

Authors:  Raphael Sznitman; Carlos Becker; Pascal Fua
Journal:  Med Image Comput Comput Assist Interv       Date:  2014

6.  circlize Implements and enhances circular visualization in R.

Authors:  Zuguang Gu; Lei Gu; Roland Eils; Matthias Schlesner; Benedikt Brors
Journal:  Bioinformatics       Date:  2014-06-14       Impact factor: 6.937

7.  EndoNet: A Deep Architecture for Recognition Tasks on Laparoscopic Videos.

Authors:  Andru P Twinanda; Sherif Shehata; Didier Mutter; Jacques Marescaux; Michel de Mathelin; Nicolas Padoy
Journal:  IEEE Trans Med Imaging       Date:  2016-07-22       Impact factor: 10.048

8.  UpSet: Visualization of Intersecting Sets.

Authors:  Alexander Lex; Nils Gehlenborg; Hendrik Strobelt; Romain Vuillemot; Hanspeter Pfister
Journal:  IEEE Trans Vis Comput Graph       Date:  2014-12       Impact factor: 4.579

  8 in total
  3 in total

1.  Articulated Multi-Instrument 2-D Pose Estimation Using Fully Convolutional Networks.

Authors:  Xiaofei Du; Thomas Kurmann; Ping-Lin Chang; Maximilian Allan; Sebastien Ourselin; Raphael Sznitman; John D Kelly; Danail Stoyanov
Journal:  IEEE Trans Med Imaging       Date:  2018-05       Impact factor: 10.048

2.  A contextual detector of surgical tools in laparoscopic videos using deep learning.

Authors:  Babak Namazi; Ganesh Sankaranarayanan; Venkat Devarajan
Journal:  Surg Endosc       Date:  2021-02-08       Impact factor: 4.584

3.  Can surgical simulation be used to train detection and classification of neural networks?

Authors:  Odysseas Zisimopoulos; Evangello Flouty; Mark Stacey; Sam Muscroft; Petros Giataganas; Jean Nehme; Andre Chow; Danail Stoyanov
Journal:  Healthc Technol Lett       Date:  2017-09-14
  3 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.