Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis

Quirine E.W. van der Zander; Ramon-Michel Schreuder; Roger Fonollà; Thom Scheeve; Fons van der Sommen; Bjorn Winkens; Patrick Aepli; Bu'Hussain Hayee; Andreas-Bernd Pischel; Milan Stefanovič; Sharmila Subramaniam; Pradeep Bhandari; Peter H.N. de With; Ad A.M. Masclee; Erik J. Schoon

doi:10.1055/a-1343-1597

Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis

Quirine E.W. van der Zander (Corresponding author), Ramon-Michel Schreuder, Roger Fonollà, Thom Scheeve, Fons van der Sommen, Bjorn Winkens, Patrick Aepli, Bu'Hussain Hayee, Andreas-Bernd Pischel, Milan Stefanovič, Sharmila Subramaniam, Pradeep Bhandari, Peter H.N. de With, Ad A.M. Masclee, Erik J. Schoon

Research output: Contribution to journal › Article › Academic › peer-review

22 Citations (Scopus)

Abstract

BACKGROUND: Optical diagnosis of colorectal polyps remains challenging. Image-enhancement techniques such as narrow-band imaging and blue-light imaging (BLI) can improve optical diagnosis. We developed and prospectively validated a computer-aided diagnosis system (CADx) using high-definition white-light (HDWL) and BLI images, and compared the system with the optical diagnosis of expert and novice endoscopists.

METHODS: CADx characterized colorectal polyps by exploiting artificial neural networks. Six experts and 13 novices optically diagnosed 60 colorectal polyps based on intuition. After 4 weeks, the same set of images was permuted and optically diagnosed using the BLI Adenoma Serrated International Classification (BASIC).

RESULTS: CADx had a diagnostic accuracy of 88.3 % using HDWL images and 86.7 % using BLI images. The overall diagnostic accuracy combining HDWL and BLI (multimodal imaging) was 95.0 %, which was significantly higher than that of experts (81.7 %, P = 0.03) and novices (66.7 %, P < 0.001). Sensitivity was also higher for CADx (95.6 % vs. 61.1 % and 55.4 %), whereas specificity was higher for experts compared with CADx and novices (95.6 % vs. 93.3 % and 93.2 %). For endoscopists, diagnostic accuracy did not increase when using BASIC, either for experts (intuition 79.5 % vs. BASIC 81.7 %, P = 0.14) or for novices (intuition 66.7 % vs. BASIC 66.5 %, P = 0.95).

CONCLUSION: CADx had a significantly higher diagnostic accuracy than experts and novices for the optical diagnosis of colorectal polyps. Multimodal imaging, incorporating both HDWL and BLI, improved the diagnostic accuracy of CADx. BASIC did not increase the diagnostic accuracy of endoscopists compared with intuitive optical diagnosis.

Original language	English
Pages (from-to)	1219-1226
Number of pages	8
Journal	Endoscopy
Volume	53
Issue number	12
Early online date	23 Dec 2020
DOIs	https://doi.org/10.1055/a-1343-1597
Publication status	Published - 1 Dec 2021

Bibliographical note

Funding Information:
M. Stefanovic has received speaker fees from Fujifilm Inc. S.Subrama-niam has received speaker fees from Fujifilm Inc. P. Bhandari has received research grants and speaker fees from Olympus, Fujifilm, Pen-tax, 3-D matrix, and Boston Scientific. A.A.M. Masclee has received a health care efficiency grant from ZON MW (Organization for Health Research and Development, The Netherlands), an unrestricted research grant from Will Pharma S.A., Research funding from Allegan and Grünenthal. He has also provided scientific advice to Bayer, Kyo-wa Kirin, and Takeda, and received a research grant from Pentax Europe GmBH and the Dutch Cancer Society. E.J. Schoon has received speaker fees and financial support to conduct research from Fujifilm Inc. All other authors declare that they have no conflicts of interest.

Publisher Copyright:
© 2021 American Institute of Physics Inc.. All rights reserved.

Access to Document

10.1055/a-1343-1597

Cite this

van der Zander, Q. E. W., Schreuder, R.-M., Fonollà, R., Scheeve, T., van der Sommen, F., Winkens, B., Aepli, P., Hayee, BH., Pischel, A.-B., Stefanovič, M., Subramaniam, S., Bhandari, P., de With, P. H. N., Masclee, A. A. M., & Schoon, E. J. (2021). Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis. Endoscopy, 53(12), 1219-1226. https://doi.org/10.1055/a-1343-1597

@article{9bebd6c50c3445508f9c5f412b79e12e,

title = "Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis",

abstract = "BACKGROUND: Optical diagnosis of colorectal polyps remains challenging. Image-enhancement techniques such as narrow-band imaging and blue-light imaging (BLI) can improve optical diagnosis. We developed and prospectively validated a computer-aided diagnosis system (CADx) using high-definition white-light (HDWL) and BLI images, and compared the system with the optical diagnosis of expert and novice endoscopists.METHODS: CADx characterized colorectal polyps by exploiting artificial neural networks. Six experts and 13 novices optically diagnosed 60 colorectal polyps based on intuition. After 4 weeks, the same set of images was permuted and optically diagnosed using the BLI Adenoma Serrated International Classification (BASIC).RESULTS: CADx had a diagnostic accuracy of 88.3 % using HDWL images and 86.7 % using BLI images. The overall diagnostic accuracy combining HDWL and BLI (multimodal imaging) was 95.0 %, which was significantly higher than that of experts (81.7 %, P = 0.03) and novices (66.7 %, P < 0.001). Sensitivity was also higher for CADx (95.6 % vs. 61.1 % and 55.4 %), whereas specificity was higher for experts compared with CADx and novices (95.6 % vs. 93.3 % and 93.2 %). For endoscopists, diagnostic accuracy did not increase when using BASIC, either for experts (intuition 79.5 % vs. BASIC 81.7 %, P = 0.14) or for novices (intuition 66.7 % vs. BASIC 66.5 %, P = 0.95). CONCLUSION: CADx had a significantly higher diagnostic accuracy than experts and novices for the optical diagnosis of colorectal polyps. Multimodal imaging, incorporating both HDWL and BLI, improved the diagnostic accuracy of CADx. BASIC did not increase the diagnostic accuracy of endoscopists compared with intuitive optical diagnosis.",

author = "{van der Zander}, {Quirine E.W.} and Ramon-Michel Schreuder and Roger Fonoll{\`a} and Thom Scheeve and {van der Sommen}, Fons and Bjorn Winkens and Patrick Aepli and Bu'Hussain Hayee and Andreas-Bernd Pischel and Milan Stefanovi{\v c} and Sharmila Subramaniam and Pradeep Bhandari and {de With}, {Peter H.N.} and Masclee, {Ad A.M.} and Schoon, {Erik J.}",

note = "Funding Information: M. Stefanovic has received speaker fees from Fujifilm Inc. S.Subrama-niam has received speaker fees from Fujifilm Inc. P. Bhandari has received research grants and speaker fees from Olympus, Fujifilm, Pen-tax, 3-D matrix, and Boston Scientific. A.A.M. Masclee has received a health care efficiency grant from ZON MW (Organization for Health Research and Development, The Netherlands), an unrestricted research grant from Will Pharma S.A., Research funding from Allegan and Gr{\"u}nenthal. He has also provided scientific advice to Bayer, Kyo-wa Kirin, and Takeda, and received a research grant from Pentax Europe GmBH and the Dutch Cancer Society. E.J. Schoon has received speaker fees and financial support to conduct research from Fujifilm Inc. All other authors declare that they have no conflicts of interest. Publisher Copyright: {\textcopyright} 2021 American Institute of Physics Inc.. All rights reserved.",

year = "2021",

month = dec,

day = "1",

doi = "10.1055/a-1343-1597",

language = "English",

volume = "53",

pages = "1219--1226",

journal = "Endoscopy",

issn = "0013-726X",

publisher = "Georg Thieme Verlag KG",

number = "12",

}

van der Zander, QEW, Schreuder, R-M, Fonollà, R, Scheeve, T , van der Sommen, F, Winkens, B, Aepli, P, Hayee, BH, Pischel, A-B, Stefanovič, M, Subramaniam, S, Bhandari, P, de With, PHN, Masclee, AAM & Schoon, EJ 2021, 'Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis', Endoscopy, vol. 53, no. 12, pp. 1219-1226. https://doi.org/10.1055/a-1343-1597

Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis. / van der Zander, Quirine E.W. (Corresponding author); Schreuder, Ramon-Michel; Fonollà, Roger et al.
In: Endoscopy, Vol. 53, No. 12, 01.12.2021, p. 1219-1226.

Research output: Contribution to journal › Article › Academic › peer-review

TY - JOUR

T1 - Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis

AU - van der Zander, Quirine E.W.

AU - Schreuder, Ramon-Michel

AU - Fonollà, Roger

AU - Scheeve, Thom

AU - van der Sommen, Fons

AU - Winkens, Bjorn

AU - Aepli, Patrick

AU - Hayee, Bu'Hussain

AU - Pischel, Andreas-Bernd

AU - Stefanovič, Milan

AU - Subramaniam, Sharmila

AU - Bhandari, Pradeep

AU - de With, Peter H.N.

AU - Masclee, Ad A.M.

AU - Schoon, Erik J.

N1 - Funding Information: M. Stefanovic has received speaker fees from Fujifilm Inc. S.Subrama-niam has received speaker fees from Fujifilm Inc. P. Bhandari has received research grants and speaker fees from Olympus, Fujifilm, Pen-tax, 3-D matrix, and Boston Scientific. A.A.M. Masclee has received a health care efficiency grant from ZON MW (Organization for Health Research and Development, The Netherlands), an unrestricted research grant from Will Pharma S.A., Research funding from Allegan and Grünenthal. He has also provided scientific advice to Bayer, Kyo-wa Kirin, and Takeda, and received a research grant from Pentax Europe GmBH and the Dutch Cancer Society. E.J. Schoon has received speaker fees and financial support to conduct research from Fujifilm Inc. All other authors declare that they have no conflicts of interest. Publisher Copyright: © 2021 American Institute of Physics Inc.. All rights reserved.

PY - 2021/12/1

Y1 - 2021/12/1

N2 - BACKGROUND: Optical diagnosis of colorectal polyps remains challenging. Image-enhancement techniques such as narrow-band imaging and blue-light imaging (BLI) can improve optical diagnosis. We developed and prospectively validated a computer-aided diagnosis system (CADx) using high-definition white-light (HDWL) and BLI images, and compared the system with the optical diagnosis of expert and novice endoscopists.METHODS: CADx characterized colorectal polyps by exploiting artificial neural networks. Six experts and 13 novices optically diagnosed 60 colorectal polyps based on intuition. After 4 weeks, the same set of images was permuted and optically diagnosed using the BLI Adenoma Serrated International Classification (BASIC).RESULTS: CADx had a diagnostic accuracy of 88.3 % using HDWL images and 86.7 % using BLI images. The overall diagnostic accuracy combining HDWL and BLI (multimodal imaging) was 95.0 %, which was significantly higher than that of experts (81.7 %, P = 0.03) and novices (66.7 %, P < 0.001). Sensitivity was also higher for CADx (95.6 % vs. 61.1 % and 55.4 %), whereas specificity was higher for experts compared with CADx and novices (95.6 % vs. 93.3 % and 93.2 %). For endoscopists, diagnostic accuracy did not increase when using BASIC, either for experts (intuition 79.5 % vs. BASIC 81.7 %, P = 0.14) or for novices (intuition 66.7 % vs. BASIC 66.5 %, P = 0.95). CONCLUSION: CADx had a significantly higher diagnostic accuracy than experts and novices for the optical diagnosis of colorectal polyps. Multimodal imaging, incorporating both HDWL and BLI, improved the diagnostic accuracy of CADx. BASIC did not increase the diagnostic accuracy of endoscopists compared with intuitive optical diagnosis.

AB - BACKGROUND: Optical diagnosis of colorectal polyps remains challenging. Image-enhancement techniques such as narrow-band imaging and blue-light imaging (BLI) can improve optical diagnosis. We developed and prospectively validated a computer-aided diagnosis system (CADx) using high-definition white-light (HDWL) and BLI images, and compared the system with the optical diagnosis of expert and novice endoscopists.METHODS: CADx characterized colorectal polyps by exploiting artificial neural networks. Six experts and 13 novices optically diagnosed 60 colorectal polyps based on intuition. After 4 weeks, the same set of images was permuted and optically diagnosed using the BLI Adenoma Serrated International Classification (BASIC).RESULTS: CADx had a diagnostic accuracy of 88.3 % using HDWL images and 86.7 % using BLI images. The overall diagnostic accuracy combining HDWL and BLI (multimodal imaging) was 95.0 %, which was significantly higher than that of experts (81.7 %, P = 0.03) and novices (66.7 %, P < 0.001). Sensitivity was also higher for CADx (95.6 % vs. 61.1 % and 55.4 %), whereas specificity was higher for experts compared with CADx and novices (95.6 % vs. 93.3 % and 93.2 %). For endoscopists, diagnostic accuracy did not increase when using BASIC, either for experts (intuition 79.5 % vs. BASIC 81.7 %, P = 0.14) or for novices (intuition 66.7 % vs. BASIC 66.5 %, P = 0.95). CONCLUSION: CADx had a significantly higher diagnostic accuracy than experts and novices for the optical diagnosis of colorectal polyps. Multimodal imaging, incorporating both HDWL and BLI, improved the diagnostic accuracy of CADx. BASIC did not increase the diagnostic accuracy of endoscopists compared with intuitive optical diagnosis.

UR - http://www.scopus.com/inward/record.url?scp=85102513459&partnerID=8YFLogxK

U2 - 10.1055/a-1343-1597

DO - 10.1055/a-1343-1597

M3 - Article

C2 - 33368056

AN - SCOPUS:85102513459

SN - 0013-726X

VL - 53

SP - 1219

EP - 1226

JO - Endoscopy

JF - Endoscopy

IS - 12

ER -

Optical diagnosis of colorectal polyp images using a newly developed computer-aided diagnosis system (CADx) compared with intuitive optical diagnosis

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this