Identifying key mechanisms leading to visual recognition errors for missed colorectal polyps using eye-tracking technology

Article

Ahmad, O., Mazomenos, E., Chadebecq, F., Kader, R., Hussein, M., Haidry, R., Puyal, J., Brandao, P., Toth, D. and Mountney, P. 2023. Identifying key mechanisms leading to visual recognition errors for missed colorectal polyps using eye-tracking technology. Journal of Gastroenterology and Hepatology. 38 (5), pp. 768-774. https://doi.org/10.1111/jgh.16127
TypeArticle
TitleIdentifying key mechanisms leading to visual recognition errors for missed colorectal polyps using eye-tracking technology
AuthorsAhmad, O., Mazomenos, E., Chadebecq, F., Kader, R., Hussein, M., Haidry, R., Puyal, J., Brandao, P., Toth, D. and Mountney, P.
Abstract

Background and Aim
Lack of visual recognition of colorectal polyps may lead to interval cancers. The mechanisms contributing to perceptual variation, particularly for subtle and advanced colorectal neoplasia, have scarcely been investigated. We aimed to evaluate visual recognition errors and provide novel mechanistic insights.

Methods
Eleven participants (seven trainees and four medical students) evaluated images from the UCL polyp perception dataset, containing 25 polyps, using eye-tracking equipment. Gaze errors were defined as those where the lesion was not observed according to eye-tracking technology. Cognitive errors occurred when lesions were observed but not recognized as polyps by participants. A video study was also performed including 39 subtle polyps, where polyp recognition performance was compared with a convolutional neural network.

Results
Cognitive errors occurred more frequently than gaze errors overall (65.6%), with a significantly higher proportion in trainees (P = 0.0264). In the video validation, the convolutional neural network detected significantly more polyps than trainees and medical students, with per-polyp sensitivities of 79.5%, 30.0%, and 15.4%, respectively.

Conclusions
Cognitive errors were the most common reason for visual recognition errors. The impact of interventions such as artificial intelligence, particularly on different types of perceptual errors, needs further investigation including potential effects on learning curves. To facilitate future research, a publicly accessible visual perception colonoscopy polyp database was created.

Keywordsartificial intelligence; colonic polyps; colonoscopy; colorectal cancer
Sustainable Development Goals3 Good health and well-being
Middlesex University ThemeHealth & Wellbeing
PublisherWiley
JournalJournal of Gastroenterology and Hepatology
ISSN0815-9319
Electronic1440-1746
Publication dates
Online18 Jan 2023
PrintMay 2023
Publication process dates
Accepted14 Jan 2023
Deposited12 Feb 2024
Output statusPublished
Publisher's version
License
File Access Level
Open
Digital Object Identifier (DOI)https://doi.org/10.1111/jgh.16127
Web of Science identifierWOS:000920025000001
Permalink -

https://repository.mdx.ac.uk/item/z6582

Log in to edit

Download files

Publisher's version
J of Gastro and Hepatol - 2023 - Ahmad - Identifying key mechanisms leading to visual recognition errors for missed.pdf
License: CC BY 4.0
File access level: Open

  • 30
    total views
  • 8
    total downloads
  • 2
    views this month
  • 1
    downloads this month

Export as

Related outputs

Artificial intelligence and automation in endoscopy and surgery
Chadebecq, F., Lovat, L. and Stoyanov, D. 2023. Artificial intelligence and automation in endoscopy and surgery. Nature Reviews Gastroenterology and Hepatology. 20 (3), pp. 171-182. https://doi.org/10.1038/s41575-022-00701-y
SuperPoint features in endoscopy
Leon Barbed, O., Chadebecq, F., Morlana, J., Montiel, J. and Murillo, A. 2022. SuperPoint features in endoscopy. Manfredi, L., Ahmadi, S., Bronstein, M., Kazi, A., Lomanto, D., Mathew, A., Magerand, L., Mullakaeva, K., Papiez, B., Taylor, R. and Trucco, E. (ed.) 1st MICCAI Workshop on Imaging Systems for GI Endoscopy. Singapore 18 - 18 Sep 2022 Cham Springer. https://doi.org/10.1007/978-3-031-21083-9_5
P19 Structure-from-motion analysis may generate an accurate automated bowel preparation score
Chadebecq, F., Mountney, P., Ahmad, O., Kader, R., Lovat, L. and Stoyanov, D. 2021. P19 Structure-from-motion analysis may generate an accurate automated bowel preparation score. BSG Campus 2021. Online 21 - 29 Jan 2021 BMJ Publishing Group Ltd. https://doi.org/gutjnl-2020-bsgcampus.94
Refractive two-view reconstruction for underwater 3D vision
Chadebecq, F., Vasconcelos, F., Lacher, R., Maneas, E., Desjardins, A., Ourselin, S., Vercauteren, T. and Stoyanov, D. 2020. Refractive two-view reconstruction for underwater 3D vision. International Journal of Computer Vision. 128 (5), pp. 1101-1117. https://doi.org/10.1007/s11263-019-01218-9
Computer vision in the surgical operating room
Chadebecq, F., Vasconcelos, F., Mazomenos, E. and Stoyanov, D. 2020. Computer vision in the surgical operating room. Visceral Medicine. 36 (6), pp. 456-462. https://doi.org/10.1159/000511934
Automated pick-up of suturing needles for robotic surgical assistance
D'Ettorre, C., Dwyer, G., Du, X., Chadebecq, F., Vasconcelos, F., De Momi, E. and Stoyanov, D. 2018. Automated pick-up of suturing needles for robotic surgical assistance. 2018 IEEE International Conference on Robotics and Automation (ICRA). Brisbane, QLD, Australia 21 - 25 May 2018 IEEE. pp. 1370-1377 https://doi.org/10.1109/ICRA.2018.8461200
Monocular 3D reconstruction of the colon using CNNs trained on synthetic data
Rau, A., Chadebecq, F., Stoyanov, D. and Riordan, P. 2018. Monocular 3D reconstruction of the colon using CNNs trained on synthetic data. 8th Joint Workshop on New Technologies for Computer/Robot Assisted Surgery. London, UK 10 - 11 Sep 2018 CRAS.
Adjoint transformation algorithm for hand-eye calibration with applications in robotic assisted surgery
Pachtrachai, K., Vasconcelos, F., Chadebecq, F., Allan, M., Hailes, S., Pawar, V. and Stoyanov, D. 2018. Adjoint transformation algorithm for hand-eye calibration with applications in robotic assisted surgery. Annals of Biomedical Engineering. 46 (10), pp. 1606-1620. https://doi.org/10.1007/s10439-018-2097-4
Medical-grade sterilizable target for fluid-immersed fetoscope optical distortion calibration
Nikitichev, D., Shakir, D., Chadebecq, F., Tella, M., Deprest, J., Stoyanov, D., Ourselin, S. and Vercauteren, T. 2017. Medical-grade sterilizable target for fluid-immersed fetoscope optical distortion calibration. Journal of Visualized Experiments. 120, p. e55298. https://doi.org/10.3791/55298
Corner-based geometric calibration of multi-focus plenoptic cameras
Nousias, S., Chadebecq, F., Pichat, J., Keane, P., Ourselin, S. and Bergeles, C. 2017. Corner-based geometric calibration of multi-focus plenoptic cameras. 2017 IEEE International Conference on Computer Vision (ICCV). Venice, Italy 22 - 29 Oct 2017 IEEE. pp. 957-965 https://doi.org/10.1109/ICCV.2017.109
Refractive structure-from-motion through a flat refractive interface
Chadebecq, F., Vasconcelos, F., Dwyer, G., Lacher, R., Ourselin, S., Vercauteren, T. and Stoyanov, D. 2017. Refractive structure-from-motion through a flat refractive interface. 2017 IEEE International Conference on Computer Vision (ICCV). Venice, Italy 22 - 29 Oct 2017 IEEE. pp. 5325-5333 https://doi.org/10.1109/ICCV.2017.568
A continuum robot and control interface for surgical assist in fetoscopic interventions
Dwyer, G., Chadebecq, F., Amo, M., Bergeles, C., Maneas, E., Pawar, V., Poorten, E., Deprest, J., Ourselin, S., De Coppi, P., Vercauteren, T. and Stoyanov, D. 2017. A continuum robot and control interface for surgical assist in fetoscopic interventions. IEEE Robotics and Automation Letters. 2 (3), pp. 1656-1663. https://doi.org/10.1109/LRA.2017.2679902
Cooperative control with distal manipulation for fetoscopic laser photocoagulation
Dwyer, G., Bergeles, C., Chadebecq, F., Pawar, V., Vander Poorten, E., Ourselin, S., Deprest, J., De Coppi, P., Vercauteren, T. and Stoyanov, D. 2016. Cooperative control with distal manipulation for fetoscopic laser photocoagulation. 6th Joint Workshop on New Technologies for Computer/Robot Assisted Surgery. Pisa, Italy 12 - 14 Sep 2016 Leuven, Belgium KU Leuven. pp. 34-35
Real-time mosaicing of fetoscopic videos using SIFT
Daga, P., Chadebecq, F., Shakir, D., Herrera, L., Tella, M., Dwyer, G., David, A., Deprest, J., Stoyanov, D., Vercauteren, T. and Ourselin, S. 2016. Real-time mosaicing of fetoscopic videos using SIFT. SPIE Medical Imaging 2016. San Diego, California, United States 27 Feb - 03 Mar 2016 SPIE. https://doi.org/10.1117/12.2217172
A combined EM and visual tracking probabilistic model for robust mosaicking: application to fetoscopy
Tella-Amo, M., Daga, P., Chadebecq, F., Thompson, S., Shakir, D., Dwyer, G., Wimalasundera, R., Deprest, J., Stoyanov, D. and Vercauteren, T. 2016. A combined EM and visual tracking probabilistic model for robust mosaicking: application to fetoscopy. 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). Las Vegas, NV, USA 26 Jun - 01 Jul 2016 IEEE. https://doi.org/10.1109/CVPRW.2016.72
Shape-from-template
Bartoli, A., Gerard, Y., Chadebecq, F., Collins, T. and Pizarro, D. 2015. Shape-from-template. IEEE Transactions on Pattern Analysis and Machine Intelligence. 37 (10), pp. 2099-2118. https://doi.org/10.1109/TPAMI.2015.2392759
How big is this neoplasia? Live colonoscopic size measurement using the Infocus-Breakpoint
Chadebecq, F., Tilmant, C. and Bartoli, A. 2015. How big is this neoplasia? Live colonoscopic size measurement using the Infocus-Breakpoint. Medical Image Analysis. 19 (1), pp. 58-74. https://doi.org/10.1016/j.media.2014.09.002
Estimation de l'échelle absolue par vision passive monofocale et application à la mesure 3D de néoplasies en imagerie coloscopique
Chadebecq, F. 2015. Estimation de l'échelle absolue par vision passive monofocale et application à la mesure 3D de néoplasies en imagerie coloscopique. Thesis Blaise Pascal University
Practical dry calibration with medium adaptation for fluid-immersed endoscopy
Chadebecq, F., Vercauteren, T., Wimalasundera, R., Attilakos, G., David, A., Deprest, J., Ourselin, S. and Stoyanov, D. 2015. Practical dry calibration with medium adaptation for fluid-immersed endoscopy. Hamlyn Symposium on Medical Robotics 2015. London, UK 20 - 23 Jun 2015
Multimodal and multimedia image analysis and collaborative networking for digestive endoscopy
d’Orazio, L., Bartoli, A., Baetz, A., Beorchia, S., Calvary, G., Chabane, Y., Chadebecq, F., Collins, T., Laurillau, Y. and Martins-Baltar, L. 2014. Multimodal and multimedia image analysis and collaborative networking for digestive endoscopy. IRBM. 35 (2), pp. 88-93.
Enhanced imaging colonoscopy facilitates dense motion-based 3D reconstruction
Alcantarilla, P., Bartoli, A., Chadebecq, F., Tilmant, C. and Lepilliez, V. 2013. Enhanced imaging colonoscopy facilitates dense motion-based 3D reconstruction. 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Osaka, Japan 03 - 07 Jul 2013 IEEE. pp. 7346-7349 https://doi.org/10.1109/EMBC.2013.6611255
Estimation de l'échelle des néoplasies en coloscopie par détection de la profondeur de défocalisation
Chadebecq, F., Tilmant, C. and Bartoli, A. 2013. Estimation de l'échelle des néoplasies en coloscopie par détection de la profondeur de défocalisation. ORASIS 2013 - Journées francophones des jeunes chercheurs en vision par ordinateur. Cluny, France 10 - 14 Jun 2013 HAL Open Science.
Using the Infocus-Breakpoint to estimate the scale of neoplasia in colonoscopy
Chadebecq, F., Tilmant, C. and Bartoli, A. 2013. Using the Infocus-Breakpoint to estimate the scale of neoplasia in colonoscopy. 2013 IEEE 10th International Symposium on Biomedical Imaging. San Francisco, CA, USA 07 - 11 Apr 2013 IEEE. pp. 354-357 https://doi.org/10.1109/ISBI.2013.6556485
Estimation de l'échelle en coloscopie monoculaire par quantification du flou optique: étude de faisabilité
Chadebecq, F., Tilmant, C., Julien, P., Collins, T. and Bartoli, A. 2012. Estimation de l'échelle en coloscopie monoculaire par quantification du flou optique: étude de faisabilité. Reconnaissance des Formes et Intelligence Artificielle 2012. Lyon, France 24 - 27 Jan 2012 HAL Open Science.
Measuring the size of neoplasia in colonoscopy using depth-from-defocus
Chadebecq, F., Tilmant, C. and Bartoli, A. 2012. Measuring the size of neoplasia in colonoscopy using depth-from-defocus. 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society. San Diego, CA, USA 28 Aug - 01 Sep 2012 IEEE. pp. 1478-1481 https://doi.org/10.1109/EMBC.2012.6346220
On template-based reconstruction from a single view: analytical solutions and proofs of well-posedness for developable, isometric and conformal surfaces
Bartoli, A., Gerard, Y., Chadebecq, F. and Collins, T. 2012. On template-based reconstruction from a single view: analytical solutions and proofs of well-posedness for developable, isometric and conformal surfaces. 2012 IEEE Conference on Computer Vision and Pattern Recognition. Providence, RI, USA 16 - 21 Jun 2012 IEEE. pp. 2026-2033 https://doi.org/10.1109/CVPR.2012.6247906