A multimodal (FACILE) classification for optical diagnosis of inflammatory bowel disease associated neoplasia

Marietta Iacucci, Kenneth McQuaid, X. Sean Gui, Yasushi Iwao, Brendan C. Lethebe, Mark Lowerison, Takayuki Matsumoto, Uday N. Shivaji, Samuel C.L. Smith, Venkataraman Subramanian, Toshio Uraoka, Silvia Sanduleanu, Subrata Ghosh, Ralf Kiesslich

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Background Characterization of colonic lesions in inflammatory bowel disease (IBD) remains challenging. We developed an endoscopic classification of visual characteristics to identify colitis-associated neoplasia using multimodal advanced endoscopic imaging (Frankfurt Advanced Chromoendoscopic IBD LEsions [FACILE] classification). Methods The study was conducted in three phases: 1) development-an expert panel defined endoscopic signs and predictors of dysplasia in IBD and, using multivariable logistic regression created the FACILE classification; 2) validation-using 60 IBD lesions from an image library, two assessments of diagnostic accuracy for neoplasia were performed and interobserver agreement between experts using FACILE was determined; 3) reproducibility-the reproducibility of the FACILE classification was tested in gastroenterologists, trainees, and junior doctors after completion of a training module. Results The experts initially selected criteria such as morphology, color, surface, vessel architecture, signs of inflammation, and lesion border. Multivariable logistic regression confirmed that nonpolypoid lesion, irregular vessel architecture, irregular surface pattern, and signs of inflammation within the lesion were predictors of dysplasia. Area under the curve of this logistic model using a bootstrapped estimate was 0.76 (0.73-0.78). The training module resulted in improved accuracy and kappa agreement in all nonexperts, though in trainees and junior doctors the kappa agreement was still moderate and poor, respectively. Conclusion We developed, validated, and demonstrated reproducibility of a new endoscopic classification (FACILE) for the diagnosis of dysplasia in IBD using all imaging modalities. Flat shape, irregular surface and vascular patterns, and signs of inflammation predicted dysplasia. The diagnostic performance of all nonexpert participants improved after a training module.

Original languageEnglish
Pages (from-to)133-141
Number of pages9
JournalEndoscopy
Volume51
Issue number2
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Inflammatory Bowel Diseases
Logistic Models
Neoplasms
Inflammation
Colitis
Libraries
Area Under Curve
Blood Vessels
Color

ASJC Scopus subject areas

  • Gastroenterology

Cite this

Iacucci, M., McQuaid, K., Gui, X. S., Iwao, Y., Lethebe, B. C., Lowerison, M., ... Kiesslich, R. (2019). A multimodal (FACILE) classification for optical diagnosis of inflammatory bowel disease associated neoplasia. Endoscopy, 51(2), 133-141. https://doi.org/10.1055/a-0757-7759

A multimodal (FACILE) classification for optical diagnosis of inflammatory bowel disease associated neoplasia. / Iacucci, Marietta; McQuaid, Kenneth; Gui, X. Sean; Iwao, Yasushi; Lethebe, Brendan C.; Lowerison, Mark; Matsumoto, Takayuki; Shivaji, Uday N.; Smith, Samuel C.L.; Subramanian, Venkataraman; Uraoka, Toshio; Sanduleanu, Silvia; Ghosh, Subrata; Kiesslich, Ralf.

In: Endoscopy, Vol. 51, No. 2, 01.01.2019, p. 133-141.

Research output: Contribution to journalArticle

Iacucci, M, McQuaid, K, Gui, XS, Iwao, Y, Lethebe, BC, Lowerison, M, Matsumoto, T, Shivaji, UN, Smith, SCL, Subramanian, V, Uraoka, T, Sanduleanu, S, Ghosh, S & Kiesslich, R 2019, 'A multimodal (FACILE) classification for optical diagnosis of inflammatory bowel disease associated neoplasia', Endoscopy, vol. 51, no. 2, pp. 133-141. https://doi.org/10.1055/a-0757-7759
Iacucci, Marietta ; McQuaid, Kenneth ; Gui, X. Sean ; Iwao, Yasushi ; Lethebe, Brendan C. ; Lowerison, Mark ; Matsumoto, Takayuki ; Shivaji, Uday N. ; Smith, Samuel C.L. ; Subramanian, Venkataraman ; Uraoka, Toshio ; Sanduleanu, Silvia ; Ghosh, Subrata ; Kiesslich, Ralf. / A multimodal (FACILE) classification for optical diagnosis of inflammatory bowel disease associated neoplasia. In: Endoscopy. 2019 ; Vol. 51, No. 2. pp. 133-141.
@article{0e902e09ac2b4035931c0ce0161124a7,
title = "A multimodal (FACILE) classification for optical diagnosis of inflammatory bowel disease associated neoplasia",
abstract = "Background Characterization of colonic lesions in inflammatory bowel disease (IBD) remains challenging. We developed an endoscopic classification of visual characteristics to identify colitis-associated neoplasia using multimodal advanced endoscopic imaging (Frankfurt Advanced Chromoendoscopic IBD LEsions [FACILE] classification). Methods The study was conducted in three phases: 1) development-an expert panel defined endoscopic signs and predictors of dysplasia in IBD and, using multivariable logistic regression created the FACILE classification; 2) validation-using 60 IBD lesions from an image library, two assessments of diagnostic accuracy for neoplasia were performed and interobserver agreement between experts using FACILE was determined; 3) reproducibility-the reproducibility of the FACILE classification was tested in gastroenterologists, trainees, and junior doctors after completion of a training module. Results The experts initially selected criteria such as morphology, color, surface, vessel architecture, signs of inflammation, and lesion border. Multivariable logistic regression confirmed that nonpolypoid lesion, irregular vessel architecture, irregular surface pattern, and signs of inflammation within the lesion were predictors of dysplasia. Area under the curve of this logistic model using a bootstrapped estimate was 0.76 (0.73-0.78). The training module resulted in improved accuracy and kappa agreement in all nonexperts, though in trainees and junior doctors the kappa agreement was still moderate and poor, respectively. Conclusion We developed, validated, and demonstrated reproducibility of a new endoscopic classification (FACILE) for the diagnosis of dysplasia in IBD using all imaging modalities. Flat shape, irregular surface and vascular patterns, and signs of inflammation predicted dysplasia. The diagnostic performance of all nonexpert participants improved after a training module.",
author = "Marietta Iacucci and Kenneth McQuaid and Gui, {X. Sean} and Yasushi Iwao and Lethebe, {Brendan C.} and Mark Lowerison and Takayuki Matsumoto and Shivaji, {Uday N.} and Smith, {Samuel C.L.} and Venkataraman Subramanian and Toshio Uraoka and Silvia Sanduleanu and Subrata Ghosh and Ralf Kiesslich",
year = "2019",
month = "1",
day = "1",
doi = "10.1055/a-0757-7759",
language = "English",
volume = "51",
pages = "133--141",
journal = "Endoscopy",
issn = "0013-726X",
publisher = "Georg Thieme Verlag",
number = "2",

}

TY - JOUR

T1 - A multimodal (FACILE) classification for optical diagnosis of inflammatory bowel disease associated neoplasia

AU - Iacucci, Marietta

AU - McQuaid, Kenneth

AU - Gui, X. Sean

AU - Iwao, Yasushi

AU - Lethebe, Brendan C.

AU - Lowerison, Mark

AU - Matsumoto, Takayuki

AU - Shivaji, Uday N.

AU - Smith, Samuel C.L.

AU - Subramanian, Venkataraman

AU - Uraoka, Toshio

AU - Sanduleanu, Silvia

AU - Ghosh, Subrata

AU - Kiesslich, Ralf

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Background Characterization of colonic lesions in inflammatory bowel disease (IBD) remains challenging. We developed an endoscopic classification of visual characteristics to identify colitis-associated neoplasia using multimodal advanced endoscopic imaging (Frankfurt Advanced Chromoendoscopic IBD LEsions [FACILE] classification). Methods The study was conducted in three phases: 1) development-an expert panel defined endoscopic signs and predictors of dysplasia in IBD and, using multivariable logistic regression created the FACILE classification; 2) validation-using 60 IBD lesions from an image library, two assessments of diagnostic accuracy for neoplasia were performed and interobserver agreement between experts using FACILE was determined; 3) reproducibility-the reproducibility of the FACILE classification was tested in gastroenterologists, trainees, and junior doctors after completion of a training module. Results The experts initially selected criteria such as morphology, color, surface, vessel architecture, signs of inflammation, and lesion border. Multivariable logistic regression confirmed that nonpolypoid lesion, irregular vessel architecture, irregular surface pattern, and signs of inflammation within the lesion were predictors of dysplasia. Area under the curve of this logistic model using a bootstrapped estimate was 0.76 (0.73-0.78). The training module resulted in improved accuracy and kappa agreement in all nonexperts, though in trainees and junior doctors the kappa agreement was still moderate and poor, respectively. Conclusion We developed, validated, and demonstrated reproducibility of a new endoscopic classification (FACILE) for the diagnosis of dysplasia in IBD using all imaging modalities. Flat shape, irregular surface and vascular patterns, and signs of inflammation predicted dysplasia. The diagnostic performance of all nonexpert participants improved after a training module.

AB - Background Characterization of colonic lesions in inflammatory bowel disease (IBD) remains challenging. We developed an endoscopic classification of visual characteristics to identify colitis-associated neoplasia using multimodal advanced endoscopic imaging (Frankfurt Advanced Chromoendoscopic IBD LEsions [FACILE] classification). Methods The study was conducted in three phases: 1) development-an expert panel defined endoscopic signs and predictors of dysplasia in IBD and, using multivariable logistic regression created the FACILE classification; 2) validation-using 60 IBD lesions from an image library, two assessments of diagnostic accuracy for neoplasia were performed and interobserver agreement between experts using FACILE was determined; 3) reproducibility-the reproducibility of the FACILE classification was tested in gastroenterologists, trainees, and junior doctors after completion of a training module. Results The experts initially selected criteria such as morphology, color, surface, vessel architecture, signs of inflammation, and lesion border. Multivariable logistic regression confirmed that nonpolypoid lesion, irregular vessel architecture, irregular surface pattern, and signs of inflammation within the lesion were predictors of dysplasia. Area under the curve of this logistic model using a bootstrapped estimate was 0.76 (0.73-0.78). The training module resulted in improved accuracy and kappa agreement in all nonexperts, though in trainees and junior doctors the kappa agreement was still moderate and poor, respectively. Conclusion We developed, validated, and demonstrated reproducibility of a new endoscopic classification (FACILE) for the diagnosis of dysplasia in IBD using all imaging modalities. Flat shape, irregular surface and vascular patterns, and signs of inflammation predicted dysplasia. The diagnostic performance of all nonexpert participants improved after a training module.

UR - http://www.scopus.com/inward/record.url?scp=85060791357&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060791357&partnerID=8YFLogxK

U2 - 10.1055/a-0757-7759

DO - 10.1055/a-0757-7759

M3 - Article

VL - 51

SP - 133

EP - 141

JO - Endoscopy

JF - Endoscopy

SN - 0013-726X

IS - 2

ER -