A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy

T. Scheeve, Maarten R. Struyvenberg, Wouter L. Curvers, Jeroen de Groof, Erik J. Schoon, Jacques J.G.H.M. Bergman, F. van der Sommen, P.H.N. de With

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Volumetric laser endomicroscopy (VLE) is an advanced imaging system offering a promising solution for the detection of early Barrett’s esophagus (BE) neoplasia. BE is a known precursor lesion for esophageal adenocarcinoma and is often missed during regular endoscopic surveillance of BE patients. VLE provides a circumferential scan of near-microscopic resolution of the esophageal wall up to 3-mm depth, yielding a large amount of data that is hard to interpret in real time. In a preliminary study on an automated analysis system for ex vivo VLE scans, novel quantitative image features were developed for two previously identified clinical VLE features predictive for BE neoplasia, showing promising results. This paper proposes a novel quantitative image feature for a missing third clinical VLE feature. The novel gland-based image feature called “gland statistics” (GS), is compared to several generic image analysis features and the most promising clinically-inspired feature “layer histogram” (LH). All features are evaluated on a clinical, validated data set consisting of 88 non-dysplastic BE and 34 neoplastic in vivo VLE images for eight different widely-used machine learning methods. The new clinically-inspired feature has on average superior classification accuracy (0.84 AUC) compared to the generic image analysis features (0.61 AUC), as well as comparable performance to the LH feature (0.86 AUC). Also, the LH feature achieves superior classification accuracy compared to the generic image analysis features in vivo, confirming previous ex vivo results. Combining the LH and the novel GS features provides even further improvement of the performance (0.88 AUC), showing great promise for the clinical utility of this algorithm to detect early BE neoplasia.
LanguageEnglish
Title of host publicationProceedings of SPIE
Subtitle of host publicationMedical Imaging 2019: Computer-Aided Diagnosis
EditorsKensaku Mori, Horst K. Hahn
PublisherInternational Society for Optics and Photonics
Number of pages6
Volume10950
DOIs
StatePublished - 13 Mar 2019
EventSPIE Medical Imaging 2019 - San Diego, United States
Duration: 16 Feb 201921 Feb 2019
http://spie.org/MI/entireprogram/2019-2-20?print=2&SSO=1

Conference

ConferenceSPIE Medical Imaging 2019
CountryUnited States
CitySan Diego
Period16/02/1921/02/19
Internet address

Fingerprint

Lasers
Image analysis
Statistics
Imaging systems
Learning systems

Cite this

Scheeve, T., Struyvenberg, M. R., Curvers, W. L., de Groof, J., Schoon, E. J., Bergman, J. J. G. H. M., ... de With, P. H. N. (2019). A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy. In K. Mori, & H. K. Hahn (Eds.), Proceedings of SPIE: Medical Imaging 2019: Computer-Aided Diagnosis (Vol. 10950). [10950-70] International Society for Optics and Photonics. DOI: 10.1117/12.2508244
Scheeve, T. ; Struyvenberg, Maarten R. ; Curvers, Wouter L. ; de Groof, Jeroen ; Schoon, Erik J. ; Bergman, Jacques J.G.H.M. ; van der Sommen, F. ; de With, P.H.N./ A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy. Proceedings of SPIE: Medical Imaging 2019: Computer-Aided Diagnosis. editor / Kensaku Mori ; Horst K. Hahn. Vol. 10950 International Society for Optics and Photonics, 2019.
@inproceedings{24461ce3ce614b3ab72bd40252f9663c,
title = "A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy",
abstract = "Volumetric laser endomicroscopy (VLE) is an advanced imaging system offering a promising solution for the detection of early Barrett’s esophagus (BE) neoplasia. BE is a known precursor lesion for esophageal adenocarcinoma and is often missed during regular endoscopic surveillance of BE patients. VLE provides a circumferential scan of near-microscopic resolution of the esophageal wall up to 3-mm depth, yielding a large amount of data that is hard to interpret in real time. In a preliminary study on an automated analysis system for ex vivo VLE scans, novel quantitative image features were developed for two previously identified clinical VLE features predictive for BE neoplasia, showing promising results. This paper proposes a novel quantitative image feature for a missing third clinical VLE feature. The novel gland-based image feature called “gland statistics” (GS), is compared to several generic image analysis features and the most promising clinically-inspired feature “layer histogram” (LH). All features are evaluated on a clinical, validated data set consisting of 88 non-dysplastic BE and 34 neoplastic in vivo VLE images for eight different widely-used machine learning methods. The new clinically-inspired feature has on average superior classification accuracy (0.84 AUC) compared to the generic image analysis features (0.61 AUC), as well as comparable performance to the LH feature (0.86 AUC). Also, the LH feature achieves superior classification accuracy compared to the generic image analysis features in vivo, confirming previous ex vivo results. Combining the LH and the novel GS features provides even further improvement of the performance (0.88 AUC), showing great promise for the clinical utility of this algorithm to detect early BE neoplasia.",
author = "T. Scheeve and Struyvenberg, {Maarten R.} and Curvers, {Wouter L.} and {de Groof}, Jeroen and Schoon, {Erik J.} and Bergman, {Jacques J.G.H.M.} and {van der Sommen}, F. and {de With}, P.H.N.",
year = "2019",
month = "3",
day = "13",
doi = "10.1117/12.2508244",
language = "English",
volume = "10950",
editor = "Kensaku Mori and Hahn, {Horst K.}",
booktitle = "Proceedings of SPIE",
publisher = "International Society for Optics and Photonics",

}

Scheeve, T, Struyvenberg, MR, Curvers, WL, de Groof, J, Schoon, EJ, Bergman, JJGHM, van der Sommen, F & de With, PHN 2019, A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy. in K Mori & HK Hahn (eds), Proceedings of SPIE: Medical Imaging 2019: Computer-Aided Diagnosis. vol. 10950, 10950-70, International Society for Optics and Photonics, SPIE Medical Imaging 2019, San Diego, United States, 16/02/19. DOI: 10.1117/12.2508244

A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy. / Scheeve, T.; Struyvenberg, Maarten R.; Curvers, Wouter L.; de Groof, Jeroen; Schoon, Erik J.; Bergman, Jacques J.G.H.M.; van der Sommen, F.; de With, P.H.N.

Proceedings of SPIE: Medical Imaging 2019: Computer-Aided Diagnosis. ed. / Kensaku Mori; Horst K. Hahn. Vol. 10950 International Society for Optics and Photonics, 2019. 10950-70.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy

AU - Scheeve,T.

AU - Struyvenberg,Maarten R.

AU - Curvers,Wouter L.

AU - de Groof,Jeroen

AU - Schoon,Erik J.

AU - Bergman,Jacques J.G.H.M.

AU - van der Sommen,F.

AU - de With,P.H.N.

PY - 2019/3/13

Y1 - 2019/3/13

N2 - Volumetric laser endomicroscopy (VLE) is an advanced imaging system offering a promising solution for the detection of early Barrett’s esophagus (BE) neoplasia. BE is a known precursor lesion for esophageal adenocarcinoma and is often missed during regular endoscopic surveillance of BE patients. VLE provides a circumferential scan of near-microscopic resolution of the esophageal wall up to 3-mm depth, yielding a large amount of data that is hard to interpret in real time. In a preliminary study on an automated analysis system for ex vivo VLE scans, novel quantitative image features were developed for two previously identified clinical VLE features predictive for BE neoplasia, showing promising results. This paper proposes a novel quantitative image feature for a missing third clinical VLE feature. The novel gland-based image feature called “gland statistics” (GS), is compared to several generic image analysis features and the most promising clinically-inspired feature “layer histogram” (LH). All features are evaluated on a clinical, validated data set consisting of 88 non-dysplastic BE and 34 neoplastic in vivo VLE images for eight different widely-used machine learning methods. The new clinically-inspired feature has on average superior classification accuracy (0.84 AUC) compared to the generic image analysis features (0.61 AUC), as well as comparable performance to the LH feature (0.86 AUC). Also, the LH feature achieves superior classification accuracy compared to the generic image analysis features in vivo, confirming previous ex vivo results. Combining the LH and the novel GS features provides even further improvement of the performance (0.88 AUC), showing great promise for the clinical utility of this algorithm to detect early BE neoplasia.

AB - Volumetric laser endomicroscopy (VLE) is an advanced imaging system offering a promising solution for the detection of early Barrett’s esophagus (BE) neoplasia. BE is a known precursor lesion for esophageal adenocarcinoma and is often missed during regular endoscopic surveillance of BE patients. VLE provides a circumferential scan of near-microscopic resolution of the esophageal wall up to 3-mm depth, yielding a large amount of data that is hard to interpret in real time. In a preliminary study on an automated analysis system for ex vivo VLE scans, novel quantitative image features were developed for two previously identified clinical VLE features predictive for BE neoplasia, showing promising results. This paper proposes a novel quantitative image feature for a missing third clinical VLE feature. The novel gland-based image feature called “gland statistics” (GS), is compared to several generic image analysis features and the most promising clinically-inspired feature “layer histogram” (LH). All features are evaluated on a clinical, validated data set consisting of 88 non-dysplastic BE and 34 neoplastic in vivo VLE images for eight different widely-used machine learning methods. The new clinically-inspired feature has on average superior classification accuracy (0.84 AUC) compared to the generic image analysis features (0.61 AUC), as well as comparable performance to the LH feature (0.86 AUC). Also, the LH feature achieves superior classification accuracy compared to the generic image analysis features in vivo, confirming previous ex vivo results. Combining the LH and the novel GS features provides even further improvement of the performance (0.88 AUC), showing great promise for the clinical utility of this algorithm to detect early BE neoplasia.

U2 - 10.1117/12.2508244

DO - 10.1117/12.2508244

M3 - Conference contribution

VL - 10950

BT - Proceedings of SPIE

PB - International Society for Optics and Photonics

ER -

Scheeve T, Struyvenberg MR, Curvers WL, de Groof J, Schoon EJ, Bergman JJGHM et al. A novel clinical gland feature for detection of early Barrett’s neoplasia using volumetric laser endomicroscopy. In Mori K, Hahn HK, editors, Proceedings of SPIE: Medical Imaging 2019: Computer-Aided Diagnosis. Vol. 10950. International Society for Optics and Photonics. 2019. 10950-70. Available from, DOI: 10.1117/12.2508244