The role of confocal laser endomicroscopy in assessing mucosal healing in patients with ulcerative proctitis

Background and Aims: Mucosal healing (MH) is associated with a stable course of Crohn’s disease (CD) which can be assessed by confocal laser endomicroscopy (CLE). To minimize the operator’s errors and automate assessment of CLE images, we used a deep learning (DL) model for image analysis. We hypothesized that DL combined with convolutional neural networks (CNNs) and long short-term memory (LSTM) can distinguish between normal and inflamed colonic mucosa from CLE images. Methods: The study included 54 patients, 32 with known active CD, and 22 control patients (18 CD patients with MH and four normal mucosa patients with no history of inflammatory bowel diseases). We designed and trained a deep convolutional neural network to detect active CD using 6,205 endomicroscopy images classified as active CD inflammation (3,672 images) and control mucosal healing or no inflammation (2,533 images). CLE imaging was performed on four colorectal areas and the terminal ileum. Gold standard was represented by the histopathological evaluation. The dataset was randomly split in two distinct training and testing datasets: 80% data from each patient were used for training and the remaining 20% for testing. The training dataset consists of 2,892 images with inflammation and 2,189 control images. The testing dataset consists of 780 images with inflammation and 344 control images of the colon. We used a CNN-LSTM model with four convolution layers and one LSTM layer for automatic detection of MH and CD diagnosis from CLE images. Results: CLE investigation reveals normal colonic mucosa with round crypts and inflamed mucosa with irregular crypts and tortuous and dilated blood vessels. Our method obtained a 95.3% test accuracy with a specificity of 92.78% and a sensitivity of 94.6%, with an area under each receiver operating characteristic curves of 0.98. Conclusions: Using machine learning algorithms on CLE images can successfully differentiate between inflammation and normal ileocolonic mucosa and can be used as a computer aided diagnosis for CD. Future clinical studies with a larger patient spectrum will validate our results and improve the CNN-SSTM model.

Download Full-text

Tu1078 The Role of Confocal Laser Endomicroscopy in the Management of Patients With Barrett's Esophagus: A Clinical Evidence-Based Consensus Report

Gastroenterology ◽

10.1016/s0016-5085(14)62700-1 ◽

2014 ◽

Vol 146 (5) ◽

pp. S-746

Author(s):

Kenneth K. Wang ◽

Razvan Arsenescu ◽

Helga Bertani ◽

Fabrice Caillol ◽

David L. Carr-Locke ◽

...

Keyword(s):

Barrett’S Esophagus ◽

Barrett's Esophagus ◽

Clinical Evidence ◽

Confocal Laser Endomicroscopy ◽

Confocal Laser ◽

Evidence Based ◽

Consensus Report

Download Full-text

Assessment of Endoscopic Healing by Using Advanced Technologies Reflects Histological Healing in Ulcerative Colitis

Journal of Crohn s and Colitis ◽

10.1093/ecco-jcc/jjaa056 ◽

2020 ◽

Vol 14 (9) ◽

pp. 1282-1289 ◽

Cited By ~ 2

Author(s):

Marietta Iacucci ◽

Rosanna Cannatelli ◽

Xianyong Gui ◽

Davide Zardo ◽

Alina Bazarova ◽

...

Keyword(s):

Ulcerative Colitis ◽

Roc Curves ◽

Mucosal Healing ◽

Confocal Laser Endomicroscopy ◽

Confocal Laser ◽

High Definition ◽

Endoscopic Score ◽

Virtual Chromoendoscopy ◽

Sensitivity Specificity ◽

A Prospective Study

Abstract Background Several studies have reported that ulcerative colitis [UC] patients with endoscopic mucosal healing may still have histological inflammation. We investigated the relationship between mucosal healing defined by modified PICaSSO [Paddington International Virtual ChromoendoScopy ScOre], Mayo Endoscopic Score [MES] and probe-based confocal laser endomicroscopy [pCLE] with histological indices in UC. Methods A prospective study enrolling 82 UC patients [male 66%] was conducted. High-definition colonoscopy was performed to evaluate the activity of the disease with MES assessed with High-Definition MES [HD-MES] and modified PICaSSO and targeted biopsies were taken; pCLE was then performed. Receiver operating characteristic [ROC] curves were plotted to determine the best thresholds for modified PICaSSO and pCLE scores that predicted histological healing according to the Robarts Histopathology Index [RHI] and ECAP ‘Extension, Chronicity, Activity, Plus’ histology score. Results A modified PICaSSO of ≤ 4 predicted histological healing at RHI ≤ 3, with sensitivity, specificity, accuracy and area under the ROC curve [AUROC] of 89.8%, 95.7%, 91.5% and 95.9% respectively. The sensitivity, specificity, accuracy and AUROC of HD-MES to predict histological healing by RHI were 81.4%, 95.7%, 85.4% and 92.1%, respectively. A pCLE ≤ 10 predicted histological healing with sensitivity of 94.9%, specificity of 91.3%, accuracy of 93.9% and AUROC of 96.5%. An ECAP of ≤ 10 was predicted by modified PICaSSO ≤ 4 with accuracy of 91.5% and AUROC of 95.9%. Conclusion Histological healing by RHI and ECAP is accurately predicted by HD-MES and modified virtual electronic chromoendoscopy PICaSSO, endoscopic score; and the use of pCLE did not improve the accuracy any further.

Download Full-text