A virtual chromoendoscopy artificial intelligence system to detect endoscopic and histologic activity/remission and predict clinical outcomes in ulcerative colitis.

BACKGROUND: Endoscopic and histological remission (ER, HR) are therapeutic targets in ulcerative colitis (UC). Virtual chromoendoscopy (VCE) improves endoscopic assessment and the prediction of histology; however, interobserver variability limits standardized endoscopic assessment. We aimed to develop an artificial intelligence (AI) tool to distinguish ER/activity, and predict histology and risk of flare from white-light endoscopy (WLE) and VCE videos. METHODS: 1090 endoscopic videos (67 280 frames) from 283 patients were used to develop a convolutional neural network (CNN). UC endoscopic activity was graded by experts using the Ulcerative Colitis Endoscopic Index of Severity (UCEIS) and Paddington International virtual ChromoendoScopy ScOre (PICaSSO). The CNN was trained to distinguish ER/activity on endoscopy videos, and retrained to predict HR/activity, defined according to multiple indices, and predict outcome; CNN and human agreement was measured. RESULTS: The AI system detected ER (UCEIS ≤ 1) in WLE videos with 72 % sensitivity, 87 % specificity, and an area under the receiver operating characteristic curve (AUROC) of 0.85; for detection of ER in VCE videos (PICaSSO ≤ 3), the sensitivity was 79 %, specificity 95 %, and the AUROC 0.94. The prediction of HR was similar between WLE and VCE videos (accuracies ranging from 80 % to 85 %). The model's stratification of risk of flare was similar to that of physician-assessed endoscopy scores. CONCLUSIONS: Our system accurately distinguished ER/activity and predicted HR and clinical outcome from colonoscopy videos. This is the first computer model developed to detect inflammation/healing on VCE using the PICaSSO and the first computer tool to provide endoscopic, histologic, and clinical assessment.

An International Multicenter Real-Life Prospective Study of Electronic Chromoendoscopy Score PICaSSO in Ulcerative Colitis.

BACKGROUND & AIMS: Endoscopic and histologic remission are important goals in the treatment of ulcerative colitis (UC). We investigated the correlation of the recently developed Paddington International Virtual ChromoendoScopy ScOre (PICaSSO) and other established endoscopic scores against multiple histological indices and prospectively assessed outcomes. METHODS: In this prospective multicenter international study, inflammatory activity was assessed with high-definition and virtual chromoendoscopy in the rectum and sigmoid using the Mayo Endoscopic Score (MES), UC Endoscopic Index of Severity (UCEIS), and PICaSSO. Targeted biopsies were taken for assessment using Robarts Histological Index (RHI), Nancy Histological index (NHI), ECAP (Extent, Chronicity, Activity, Plus score), Geboes, and Villanacci. Follow-up data were obtained at 6 and 12 months after colonoscopy. RESULTS: A total of 307 patients were recruited. There was strong correlation between PICaSSO and histology scores, significantly superior to correlation coefficients of MES and UCEIS with histology scores. A PICaSSO score of ≤3 detected histologic remission by RHI (≤3 + absence of neutrophils) with area under the receiver operating characteristic curve (AUROC) 0.90 (95% confidence interval [CI] 0.86-0.94) and NHI (≤1) AUROC 0.82 (95% CI 0.77-0.87). The interobserver agreement for PICaSSO was 0.88 (95% CI 0.83-0.92). At 6- and 12-months follow-up, PICaSSO score ≤3 predicted better outcomes than PICaSSO >3 (hazard ratio [HR] 0.19 [0.11-0.33] and 0.22 [0.13-0.34], respectively),} as well as PICaSSO 4-8 (HR 0.25 [0.12-0.53] and 0.22 (0.12-0.39), respectively) and similar to histologic remission. CONCLUSION: In this first real-life multicenter study, the PICaSSO score correlated strongly with multiple histological indices. Furthermore, PICaSSO score predicted specified clinical outcomes at 6 and 12 months, similar to histology. Thus, PICaSSO can be a useful endoscopic tool in the therapeutic management of UC.

Performance measures for colonoscopy in inflammatory bowel disease patients: European Society of Gastrointestinal Endoscopy (ESGE) Quality Improvement Initiative.

The European Society of Gastrointestinal Endoscopy (ESGE) presents a short list of performance measures for colonoscopy in inflammatory bowel disease (IBD) patients. Current performance measures for colonoscopy mainly focus on detecting (pre)malignant lesions. However, these performance measures are not relevant for all colonoscopy indications in IBD patients. Therefore, our aim was to provide endoscopy services across Europe and other interested countries with a tool for quality monitoring and improvement in IBD colonoscopy. Eight key performance measures and one minor performance measure were recommended for measurement and evaluation in daily endoscopy practice.

Neutrophil-only Histological Assessment of Ulcerative Colitis Correlates with Endoscopic Activity and Predicts Long-term Outcomes in a Multicentre Study.

BACKGROUNDS AND AIMS: Absence of neutrophils is the minimum standard to consider histological remission of ulcerative colitis [UC]. The PICaSSO Histological Remission Index [PHRI] is a new simple index for UC, based only on the detection of neutrophils. We evaluate PHRI's correlation with endoscopy and its prognostic value compared with other established indices. METHODS: Consecutive patients with UC underwent colonoscopy at two referral centres [Birmingham, UK, and Milan, Italy,] and were followed up for 2 years. Correlation between histology (PHRI, Nancy [NHI], and Robarts [RHI] indexes) and endoscopy (Mayo Endoscopic Score [MES], Ulcerative Colitis Endoscopic Index of Severity [UCEIS], and PICaSSO index) was calculated as Spearman coefficients. Diagnostic performance of endoscopy was assessed with receiver operating characteristic [ROC] curves and outcome stratification with Kaplan-Meier curves. RESULTS: A total of 192 patients with UC was enrolled, representing all grades of endoscopic severity. Correlation between histology and endoscopy did not differ significantly when using PHRI instead of NHI or RHI. In particular, PHRI's correlation with MES, UCEIS, and PICaSSO was 0.745, 0.718, and 0.694, respectively. Endoscopically-assessed remission reflected the absence of neutrophils [PHRI = 0] with areas under the ROC curve of 0.905, 0.906, and 0.877 for MES, UCEIS, and PICaSSO, respectively. The hazard ratio for disease flare between patients in histological activity/remission was statistically similar [p >0.05] across indexes [2.752, 2.706, and 2.871 for RHI, NHI, and PHRI, respectively]. CONCLUSION: PHRI correlates with endoscopy and stratifies risk of relapse similarly to RHI and NHI. Neutrophil-only assessment of UC is a simple yet viable alternative to established histological scores.

Computer-Aided Imaging Analysis of Probe-Based Confocal Laser Endomicroscopy With Molecular Labeling and Gene Expression Identifies Markers of Response to Biological Therapy in IBD Patients: The Endo-Omics Study.

BACKGROUND: We aimed to predict response to biologics in inflammatory bowel disease (IBD) using computerized image analysis of probe confocal laser endomicroscopy (pCLE) in vivo and assess the binding of fluorescent-labeled biologics ex vivo. Additionally, we investigated genes predictive of anti-tumor necrosis factor (TNF) response. METHODS: Twenty-nine patients (15 with Crohn's disease [CD], 14 with ulcerative colitis [UC]) underwent colonoscopy with pCLE before and 12 to 14 weeks after starting anti-TNF or anti-integrin α4ß7 therapy. Biopsies were taken for fluorescein isothiocyanate-labeled infliximab and vedolizumab staining and gene expression analysis. Computer-aided quantitative image analysis of pCLE was performed. Differentially expressed genes predictive of response were determined and validated in a public cohort. RESULTS: In vivo, vessel tortuosity, crypt morphology, and fluorescein leakage predicted response in UC (area under the receiver-operating characteristic curve [AUROC], 0.93; accuracy 85%, positive predictive value [PPV] 89%; negative predictive value [NPV] 75%) and CD (AUROC, 0.79; accuracy 80%; PPV 75%; NPV 83%) patients. Ex vivo, increased binding of labeled biologic at baseline predicted response in UC (UC) (AUROC, 83%; accuracy 77%; PPV 89%; NPV 50%) but not in Crohn's disease (AUROC 58%). A total of 325 differentially expressed genes distinguished responders from nonresponders, 86 of which fell within the most enriched pathways. A panel including ACTN1, CXCL6, LAMA4, EMILIN1, CRIP2, CXCL13, and MAPKAPK2 showed good prediction of anti-TNF response (AUROC >0.7). CONCLUSIONS: Higher mucosal binding of the drug target is associated with response to therapy in UC. In vivo, mucosal and microvascular changes detected by pCLE are associated with response to biologics in inflammatory bowel disease. Anti-TNF-responsive UC patients have a less inflamed and fibrotic state pretreatment. Chemotactic pathways involving CXCL6 or CXCL13 may be novel targets for therapy in nonresponders.

Raman spectroscopy accurately differentiates mucosal healing from non-healing and biochemical changes following biological therapy in inflammatory bowel disease.

BACKGROUND: Mucosal healing (MH) is a key treatment target in the management of inflammatory bowel disease (IBD) and is defined in endoscopic terms by the newly published PICaSSO score. Raman Spectroscopy (RS) is based on the scattering of inelastic light giving spectra that are highly specific for individual molecules. We aimed to establish spectral changes before and after treatment and whether Raman Spectroscopy is able to accurately differentiate between inflammation and MH. METHODS: Biopsies were taken for ex vivo RS analysis alongside biopsies for histological analysis from IBD patients undergoing optical diagnosis endoscopic assessment. We compared pre- vs. post-biological treatment in IBD patients and healthy controls and active vs. MH in UC and CD. For spectral analysis, we used supervised self-organising maps for separation and classification. RESULTS: A total of 23 patients (14 IBD, 9 HC) were recruited for comparison of pre- vs. post-biologic treatment and 74 IBD patients were included for the assessment of MH in IBD, giving 9700 Raman Spectra. Spectral differences were seen between pre- and post-treatment which were observed comparing MH vs. active inflammation. Reductions in intensity at 1003cm-1 and 1252cm-1 when a reduction in inflammation was seen post-treatment and when MH was present. MH was associated with an increase in intensity at 1304cm-1. The trained neural network differentiated MH from active inflammation with a sensitivity, specificity, PPV, NPV and accuracy in UC of 96.29% (sd 0.94), 95.03% (sd 1.52), 94.89% (sd 1.59), 96.33 (sd 0.97) and 95.65 (sd 0.99) and 96.19% (sd 1.46), 88% (sd 4.20), 86.60% (sd 5.39), 96.55% (sd 1.32) and 91.6% (sd 2.75) in CD respectively. CONCLUSION: We demonstrated RS can demonstrate biochemical changes following treatment of IBD and accurately differentiates MH from active inflammation in IBD and might be a future tool to personalise therapeutic management in IBD.