Cargando…

Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade

Surgical instrument detection in robot-assisted surgery videos is an import vision component for these systems. Most of the current deep learning methods focus on single-tool detection and suffer from low detection speed. To address this, the authors propose a novel frame-by-frame detection method u...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhao, Zijian, Cai, Tongbiao, Chang, Faliang, Cheng, Xiaolin
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Institution of Engineering and Technology 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6952255/
https://www.ncbi.nlm.nih.gov/pubmed/32038871
http://dx.doi.org/10.1049/htl.2019.0064
_version_ 1783486411859558400
author Zhao, Zijian
Cai, Tongbiao
Chang, Faliang
Cheng, Xiaolin
author_facet Zhao, Zijian
Cai, Tongbiao
Chang, Faliang
Cheng, Xiaolin
author_sort Zhao, Zijian
collection PubMed
description Surgical instrument detection in robot-assisted surgery videos is an import vision component for these systems. Most of the current deep learning methods focus on single-tool detection and suffer from low detection speed. To address this, the authors propose a novel frame-by-frame detection method using a cascading convolutional neural network (CNN) which consists of two different CNNs for real-time multi-tool detection. An hourglass network and a modified visual geometry group (VGG) network are applied to jointly predict the localisation. The former CNN outputs detection heatmaps representing the location of tool tip areas, and the latter performs bounding-box regression for tool tip areas on these heatmaps stacked with input RGB image frames. The authors’ method is tested on the publicly available EndoVis Challenge dataset and the ATLAS Dione dataset. The experimental results show that their method achieves better performance than mainstream detection methods in terms of detection accuracy and speed.
format Online
Article
Text
id pubmed-6952255
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher The Institution of Engineering and Technology
record_format MEDLINE/PubMed
spelling pubmed-69522552020-02-07 Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade Zhao, Zijian Cai, Tongbiao Chang, Faliang Cheng, Xiaolin Healthc Technol Lett Special Issue: Papers from the 13th Workshop on Augmented Environments for Computer Assisted Interventions Surgical instrument detection in robot-assisted surgery videos is an import vision component for these systems. Most of the current deep learning methods focus on single-tool detection and suffer from low detection speed. To address this, the authors propose a novel frame-by-frame detection method using a cascading convolutional neural network (CNN) which consists of two different CNNs for real-time multi-tool detection. An hourglass network and a modified visual geometry group (VGG) network are applied to jointly predict the localisation. The former CNN outputs detection heatmaps representing the location of tool tip areas, and the latter performs bounding-box regression for tool tip areas on these heatmaps stacked with input RGB image frames. The authors’ method is tested on the publicly available EndoVis Challenge dataset and the ATLAS Dione dataset. The experimental results show that their method achieves better performance than mainstream detection methods in terms of detection accuracy and speed. The Institution of Engineering and Technology 2019-11-26 /pmc/articles/PMC6952255/ /pubmed/32038871 http://dx.doi.org/10.1049/htl.2019.0064 Text en http://creativecommons.org/licenses/by-nc-nd/3.0/ This is an open access article published by the IET under the Creative Commons Attribution-NonCommercial-NoDerivs License (http://creativecommons.org/licenses/by-nc-nd/3.0/)
spellingShingle Special Issue: Papers from the 13th Workshop on Augmented Environments for Computer Assisted Interventions
Zhao, Zijian
Cai, Tongbiao
Chang, Faliang
Cheng, Xiaolin
Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade
title Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade
title_full Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade
title_fullStr Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade
title_full_unstemmed Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade
title_short Real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade
title_sort real-time surgical instrument detection in robot-assisted surgery using a convolutional neural network cascade
topic Special Issue: Papers from the 13th Workshop on Augmented Environments for Computer Assisted Interventions
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6952255/
https://www.ncbi.nlm.nih.gov/pubmed/32038871
http://dx.doi.org/10.1049/htl.2019.0064
work_keys_str_mv AT zhaozijian realtimesurgicalinstrumentdetectioninrobotassistedsurgeryusingaconvolutionalneuralnetworkcascade
AT caitongbiao realtimesurgicalinstrumentdetectioninrobotassistedsurgeryusingaconvolutionalneuralnetworkcascade
AT changfaliang realtimesurgicalinstrumentdetectioninrobotassistedsurgeryusingaconvolutionalneuralnetworkcascade
AT chengxiaolin realtimesurgicalinstrumentdetectioninrobotassistedsurgeryusingaconvolutionalneuralnetworkcascade