Artificial Intelligence for Disease Assessment in Inflammatory Bowel Disease: How Will it Change Our Practice?

Artificial intelligence (AI) has arrived and it will directly impact how we assess, monitor, and manage inflammatory bowel disease (IBD). Advances in the machine learning methodologies that power AI have produced astounding results for replicating expert judgment and predicting clinical outcomes, particularly in the analysis of imaging. This review will cover general concepts for AI in IBD, with descriptions of common machine learning methods, including decision trees and neural networks. Applications of AI in IBD will cover recent achievements in endoscopic image interpretation and scoring, new capabilities for cross-sectional image analysis, natural language processing for automated understanding of clinical text, and progress in AI-powered clinical decision support tools. In addition to detailing current evidence supporting the capabilities of AI for replicating expert clinical judgment, speculative commentary on how AI may advance concepts of disease activity assessment, care pathways, and pathophysiologic mechanisms of IBD will be addressed.

Serum Leucine-Rich α2 Glycoprotein: A Novel Biomarker for Transmural Inflammation in Crohn's Disease.

INTRODUCTION: Leucine-rich alpha-2 glycoprotein (LRG) is a newly studied biomarker for inflammatory diseases. This study aimed to investigate whether LRG can be used for evaluating transmural activity in patients with Crohn's disease (CD). METHODS: We performed magnetic resonance enterography (MRE) in 227 consecutive patients with CD from June 2020 to August 2021. We prospectively compared MRE findings with clinical and laboratory data including LRG. MRE was evaluated using 2 validated scoring systems, and transmural inflammation was defined as having a maximum simplified magnetic resonance index of activity (sMaRIA) score of ≥4 and a 5-point classification score of ≥9, respectively. RESULTS: The correlation between LRG and the total MRE score showed a positive correlation ( r = 0.576 for the sMaRIA score, P < 0.01, and r = 0.633 for the 5-point score, P < 0.01). Serum concentrations of LRG significantly increased as MRE scores increased ( P < 0.01). The area under the curve of LRG for a sMaRIA score of ≥4 and a 5-point score of ≥9 was 0.845 and 0.869, respectively, which was significantly higher than that of CDAI ( P < 0.01) or C-reactive protein ( P < 0.01). LRG levels of ≥14 µg/mL had a 67% sensitivity and 90% specificity for a sMaRIA score of ≥4 and a 73% sensitivity and 89% specificity for a 5-point score of ≥9. Patients with high LRG levels were also strongly associated with CD-related hospitalization, surgery, and clinical relapse compared with those with low LRG levels ( P < 0.01 for all). DISCUSSION: LRG is a highly accurate serum biomarker for detecting transmural activity in patients with CD. Results need to be validated in further multicenter studies.

Serum Leucine-Rich α2 Glycoprotein: A Novel Biomarker For Small Bowel Mucosal Activity in Crohn's Disease.

Active lesions in the small bowel (SB) have been independently associated with poorer prognoses in patients with Crohn's disease (CD)1; however, there has been a lack of accurate and convenient screening methods. Past studies have found that serum levels of the glycoprotein leucine-rich α2 glycoprotein (LRG) correlates with endoscopic activity in ulcerative colitis,2,3 and this is now available for routine clinical use as a biomarker in patients with inflammatory bowel disease in Japan. LRG has not yet been thoroughly verified in CD, and we investigated whether it can be used as a serum biomarker for detecting SB mucosal activity in patients with CD.

Mucosal healing of small intestinal stricture is associated with improved prognosis post-dilation in Crohn's disease.

BACKGROUND: Small intestinal stricture is a major cause for surgery in Crohn's disease (CD). Endoscopic balloon dilation (EBD) is performed for small intestinal strictures to avoid surgery, often repeatedly. However, factors that are associated with prognosis after EBD of small intestinal strictures remain poorly investigated. Mucosal healing is the therapeutic target in CD. We aimed to investigate the impact of mucosal healing defined by the presence of ulcers at the small intestinal stricture site on the prognosis of EBD in CD patients. METHODS: We retrospectively included patients with CD who underwent initial EBD for endoscopically impassable small intestinal strictures from January 2012 to March 2020 at a single center. The association between presence of ulcer at the stricture site and surgery after EBD was examined by Cox proportional hazards model. RESULTS: Of the 98 patients included, 63 (64.3%) had ulcer at the stricture site. 20 (31.7%) of these patients underwent surgery for the stricture in due course, whereas 4 (11.4%) of the patients without ulcer of the stricture underwent surgery. In multivariate analysis, patients with ulcer of the stricture had a significantly higher risk for surgery than those without ulcer (hazard ratio 4.84; 95% confidence interval 1.58-14.79). CONCLUSION: Mucosal healing at the stricture site indicated a favorable prognosis after EBD for small intestinal strictures in CD.

Systematic review of artificial intelligence-based image diagnosis for inflammatory bowel disease.

OBJECTIVES: Diagnosis of inflammatory bowel diseases (IBD) involves combining clinical, laboratory, endoscopic, histologic, and radiographic data. Artificial intelligence (AI) is rapidly being developed in various fields of medicine, including IBD. Because a key part in the diagnosis of IBD involves evaluating imaging data, AI is expected to play an important role in this aspect in the coming decades. We conducted a systematic literature review to highlight the current advancement of AI in diagnosing IBD from imaging data. METHODS: We performed an electronic PubMed search of the MEDLINE database for studies up to January 2022 involving IBD and AI. Studies using imaging data as input were included, and nonimaging data were excluded. RESULTS: A total of 27 studies are reviewed, including 18 studies involving endoscopic images and nine studies involving other imaging data. CONCLUSION: We highlight in this review the recent advancement of AI in diagnosing IBD from imaging data by summarizing the relevant studies, and discuss the future role of AI in clinical practice.

Development and Validation of a Deep Neural Network for Accurate Evaluation of Endoscopic Images From Patients With Ulcerative Colitis.

BACKGROUND & AIMS: There are intra- and interobserver variations in endoscopic assessment of ulcerative colitis (UC) and biopsies are often collected for histologic evaluation. We sought to develop a deep neural network system for consistent, objective, and real-time analysis of endoscopic images from patients with UC. METHODS: We constructed the deep neural network for evaluation of UC (DNUC) algorithm using 40,758 images of colonoscopies and 6885 biopsy results from 2012 patients with UC who underwent colonoscopy from January 2014 through March 2018 at a single center in Japan (the training set). We validated the accuracy of the DNUC algorithm in a prospective study of 875 patients with UC who underwent colonoscopy from April 2018 through April 2019, with 4187 endoscopic images and 4104 biopsy specimens. Endoscopic remission was defined as a UC endoscopic index of severity score of 0; histologic remission was defined as a Geboes score of 3 points or less. RESULTS: In the prospective study, the DNUC identified patients with endoscopic remission with 90.1% accuracy (95% confidence interval [CI] 89.2%-90.9%) and a kappa coefficient of 0.798 (95% CI 0.780-0.814), using findings reported by endoscopists as the reference standard. The intraclass correlation coefficient between the DNUC and the endoscopists for UC endoscopic index of severity scoring was 0.917 (95% CI 0.911-0.921). The DNUC identified patients in histologic remission with 92.9% accuracy (95% CI 92.1%-93.7%); the kappa coefficient between the DNUC and the biopsy result was 0.859 (95% CI 0.841-0.875). CONCLUSIONS: We developed a deep neural network for evaluation of endoscopic images from patients with UC that identified those in endoscopic remission with 90.1% accuracy and histologic remission with 92.9% accuracy. The DNUC can therefore identify patients in remission without the need for mucosal biopsy collection and analysis. Trial number: UMIN000031430.

Intravenous tacrolimus is a superior induction therapy for acute severe ulcerative colitis compared to oral tacrolimus.

BACKGROUND: Intravenous corticosteroid is the mainstay for managing acute severe ulcerative colitis, but one-third of patients do not respond to intravenous corticosteroid. Tacrolimus, a salvage therapy before colectomy, is usually orally administered, though its bioavailability is low compared intravenous administration. The efficacy of intravenous tacrolimus has not been widely studied. AIM: To determine the efficacy and safety of intravenous tacrolimus for the treatment of acute severe ulcerative colitis. METHODS: Eighty-seven hospitalized acute severe ulcerative colitis patients were enrolled for a prospective cohort study between 2009 and 2017. Sixty-five patients received intravenous tacrolimus and 22 received oral tacrolimus. The primary outcome was the achievement of clinical remission within 2 weeks. Relapse and colectomy incidence and adverse events were assessed at 24 weeks. RESULTS: Response rates of both treatments exceeded 50% but were not significantly different. The remission rate was higher in intravenous tacrolimus compared with oral tacrolimus. At 24 weeks, oral and intravenous tacrolimus showed similar relapse-free survival rates; however, colectomy-free survival rates were higher in intravenous tacrolimus compared with oral tacrolimus. CONCLUSIONS: Patients receiving intravenous tacrolimus achieved superior remission and colectomy-free survival rates compared with patients receiving oral tacrolimus. Safety was similar between the two treatments.

Small Bowel Healing Detected by Endoscopy in Patients With Crohn's Disease After Treatment With Antibodies Against Tumor Necrosis Factor.

BACKGROUND & AIMS: It is important to objectively assess Crohn's disease (CD) activity in patients treated with antibodies against tumor necrosis factor (anti-TNF). Detection of healing by endoscopy (endoscopic healing) associates with patient outcome, based on evidence from studies of ileocolonoscopy. We assessed endoscopic healing after treatment, based on findings from balloon-assisted enteroscopy (BAE), in patients with CD. METHODS: We performed a post-hoc analysis of data from a clinical trial from 116 patients with CD (46 with ileal and 70 with ileocolonic type) who received induction and then maintenance therapy with anti-TNF agents from January 2013 through March 2018 at a single center in Japan. We compared findings from BAE before induction therapy and then again during maintenance therapy (median 13 months later). Endoscopic healing was defined as the modified simple endoscopic score for CD below 5. We also collected data on previous treatments, makers of inflammation, and disease type. RESULTS: Before treatment, small bowel ulcerations were present in 114 patients (98%); 42 patients (60%) with ileocolonic disease had colon ulcerations. During maintenance therapy, 41/114 patients (36%) had small bowel endoscopic healing; all the patients with small bowel endoscopic healing also had colonic endoscopic healing. Colonic endoscopic healing was observed in 33/42 patients (79%). The proportion of patients with small bowel endoscopic healing was significantly lower than that of colonic endoscopic healing (P < .001). Among all patients, failure to achieve small bowel endoscopic healing was significantly associated with structuring or penetrating disease (P = .014), lack of concomitant treatment with immunomodulators (P = .015), and having received previous treatment with an anti-TNF agents (P = .018). CONCLUSIONS: In a post-hoc analysis of patients with CD treated with anti-TNF agents, we found small bowel ulcerations, detected by BAE, to be more difficult to heal than colon ulcerations.

Crohn Disease: A 5-Point MR Enterocolonography Classification Using Enteroscopic Findings.

OBJECTIVE: The objectives of our study were to establish the efficacy of a 5-point MR enterocolonography classification for assessing Crohn disease (CD) activity, compare this classification with a validated MRI score (i.e., the MR index of activity [MaRIA]), and compare both with endoscopic findings, which were assessed using the Crohn disease endoscopic index of severity (CDEIS). MATERIALS AND METHODS: Seventy (derivation cohort) and 50 (validation cohort) patients with CD were retrospectively enrolled in this study. We developed a 5-point MR enterocolonography classification that consists of visual assessments alone. MR enterocolonography results were evaluated for each bowel segment (rectum; sigmoid, descending, transverse, and ascending colon; terminal and proximal ileum; and jejunum) by one observer in the derivation phase and independently by three observers in the validation phase using the 5-point MR enterocolonography classification lexicon and MaRIA. Areas under the ROC curves (AUCs) in discriminating endoscopic deep ulcers were compared between the MR enterocolonography classification and MaRIA. Interobserver reproducibility was assessed using weighted kappa coefficients. RESULTS: The AUCs of the MR enterocolonography classification were 89.0% in the derivation phase and 88.5%, 81.0%, and 77.3% for the three observers in the validation phase. The AUCs of the MR enterocolonography classification were statistically noninferior to those of MaRIA (p < 0.001). The cross-validation accuracy was 81.9% in the derivation phase and 81.5% in the validation phase. The MR enterocolonography classification showed good reproducibility. CONCLUSION: The 5-point MR enterocolonography classification was shown to be effective for evaluating CD activity in the large and small bowel.

Predictors of mucosal healing during induction therapy in patients with acute moderate-to-severe ulcerative colitis.

BACKGROUND AND AIM: The treat-to-target strategy has emerged in ulcerative colitis management. Mucosal healing is the best target, albeit not in induction therapy of acute diseases as clinical conditions vary over a short duration. To determine the targets during induction therapy for acute ulcerative colitis, we identified markers to predict mucosal healing at 3 and 12 months of initiating the induction therapy. METHODS: This single-center prospective observational study enrolling 61 adult patients hospitalized for disease exacerbation collected the partial Mayo scores, ulcerative colitis endoscopic index of severity, fecal markers, and laboratory data (0 day, 2 weeks, and 3 and 12 months) of initiating induction therapy. RESULTS: At 2 weeks, patients with mucosal healing at 3 months had had lower partial Mayo and ulcerative colitis endoscopic index of severity scores and higher white blood cell count and total cholesterol than those without mucosal healing. At 3 months, patients with mucosal healing at 12 months had had lower partial Mayo and ulcerative colitis endoscopic index of severity scores than those without mucosal healing. A kinetic analysis demonstrated a difference in the partial Mayo scores and total cholesterol and albumin levels at 2 weeks and in the ulcerative colitis endoscopic index of severity, fecal calprotectin, and fecal immunochemical tests at 3 months between patients who achieved mucosal healing at 12 months and those who did not. CONCLUSIONS: Partial Mayo scores and total cholesterol levels act as short-term therapeutic targets during induction therapy in patients with acute ulcerative colitis. Mucosal healing at 3 months correlates to longer time mucosal healing.

Long-term effect of NUDT15 R139C on hematologic indices in inflammatory bowel disease patients treated with thiopurine.

BACKGROUND AND AIM: A missense variant of the nucleoside diphosphate-linked moiety X-type motif 15 (NUDT15) gene (R139C) predisposes Asian patients with inflammatory bowel disease (IBD) to thiopurine-induced leukopenia. This study evaluates the long-term effect of NUDT15 R139C heterozygosity on hematological parameters during thiopurine administration. METHODS: We enrolled 83 Japanese IBD patients who were on anti-tumor necrosis factor-α agents and had used thiopurine. NUDT15 R139C was genotyped by polymerase chain reaction. We retrospectively reviewed patient clinical charts to collect data on white blood cell (WBC) count, mean corpuscular volume (MCV), hemoglobin, and platelet count during the 24 months following thiopurine initiation. RESULTS: The included patients had either Crohn's disease (54; 65.1%) or ulcerative colitis (29; 34.9%). Genotyping of NUDT15 R139C identified 62 patients (74.7%) of genotype C/C and 21 (25.3%) of genotype C/T. The median dose of thiopurine was lower in the C/T group than in the C/C group after starting thiopurine. At 6 months, the mean WBC count of the C/T group became significantly lower than that of the C/C group (P = 0.008) and remained lower through the 24 months. The C/T group developed grade 2-4 leukopenia by 6 months, which persisted through 12-24 months. The mean MCV in the C/T group became higher than that of the C/C group after 3 months. CONCLUSIONS: NUDT15 R139C heterozygosity affected the WBC count and MCV for 24 months after thiopurine administration. Our results indicate that careful monitoring of leukopenia and dose adjustment are necessary throughout treatment in IBD patients heterozygous for the NUDT15 R139C.

Utility of Magnetic Resonance Enterography For Small Bowel Endoscopic Healing in Patients With Crohn's Disease.

OBJECTIVES: Small bowel (SB) endoscopic healing has not been well studied in patients with Crohn's disease (CD). This study aims to evaluate the utility of magnetic resonance (MR) enterography (MRE) for SB lesions in comparison with balloon-assisted enteroscopy (BAE) findings. METHODS: In total, 139 patients with CD in clinical-serological remission were prospectively followed after BAE and MRE procedures. We applied a modified version of the Simple Endoscopic Score for CD (SES-CD) for an endoscopic evaluation of the SB, called the Simple Endoscopic Active Score for CD (SES-CDa). We also used the MR index of activity (MaRIA) for MR evaluations. The primary end points were time to clinical relapse (CD activity index of >150 with an increase of >70 points) and serological relapse (abnormal elevation of C-reactive protein). RESULTS: Clinical and serological relapses occurred in 30 (21.6%) and 62 (44.6%) patients, respectively. SB endoscopic healing (SES-CDa<5) was observed in 76 (54.7%) patients. A multiple regression analysis showed that the lack of SB endoscopic healing was an independent risk factor for clinical relapses (hazard ratio (HR): 5.34; 95% confidence interval (CI): 2.06-13.81) and serological relapses (HR: 3.02; 95% CI: 1.65-5.51), respectively. MR ulcer healing (MaRIA score <11) demonstrated a high diagnostic accuracy (90.9%; 95% CI: 87.9-93.2%) for endoscopic healing. The kappa coefficient between BAE and MRE for longitudinal responsiveness was 0.754 (95% CI: 0.658-0.850) for clinical relapse and 0.783 (95% CI: 0.701-0.865) for serological relapse. CONCLUSIONS: SB inflammation was associated with a poor prognosis in patients with clinical-serological remission. MRE is a valid and reliable examination for SB inflammatory activity both for cross-sectional evaluations and prognostic prediction.

Prediction of disease activity of Crohn's disease through fecal calprotectin evaluated by balloon-assisted endoscopy.

BACKGROUND AND AIM: Fecal calprotectin (FC) is a useful marker for assessing the activity of intestinal inflammation. However, most studies have used ileocolonoscopy to evaluate the association of FC with intestinal inflammation, and it is not clear whether FC is useful for the evaluation of small-bowel Crohn's disease (CD). This study aimed to determine the usefulness of FC for predicting intestinal inflammation evaluated by balloon-assisted endoscopy (BAE), which can visualize the deep small intestine. METHODS: This was a cross-sectional, observational study involving 69 CD patients, 39 of whom had only small-bowel disease. The extended simplified endoscopic activity score for Crohn's disease (eSES-CD) was calculated based on the findings of BAE. Mucosal healing was defined as an eSES-CD of 0. RESULTS: In all CD patients, FC levels were correlated with the eSES-CD (r = 0.663, P < 0.001). The cutoff value to predict mucosal healing was 92 mg/kg, with a sensitivity of 94%, specificity of 88%, positive predictive value of 98%, negative predictive value of 64%, and the area under the curve of 0.91. Even in small-bowel CD patients, FC levels were correlated with the eSES-CD (r = 0.607, P < 0.001). The cutoff value was 92 mg/kg, with a sensitivity of 87%, specificity of 88%, positive predictive value of 96%, negative predictive value of 64%, and area under the curve of 0.85. CONCLUSIONS: Fecal calprotectin showed a significant correlation with the intestinal inflammation evaluated with BAE even in patients with only small intestinal disease. FC is useful for the evaluation of CD including both the small and large intestines.

Endoscopic features and genetic background of inflammatory bowel disease complicated with Takayasu arteritis.

BACKGROUND AND AIM: Takayasu arteritis (TA) is occasionally complicated with inflammatory bowel disease (IBD). This study assessed the endoscopic and genetic features of IBD complicated with TA (IBD-TA). METHODS: This study retrospectively reviewed the clinical charts of 142 TA patients (14 men and 128 women; median age 48.5 years [range, 18-97 years]). Human lymphocyte antigen (HLA) types and a single-nucleotide polymorphism rs6871626 in the IL12B gene were assessed in 101 and 81 patients with TA, respectively. RESULTS: Inflammatory bowel disease was diagnosed in 13 (9.2%) of the 142 patients. The endoscopic features of IBD-TA at initial diagnosis (n = 8) showed discontinuous and focal mucosal inflammations (n = 7, 87.5%), and only one case was diagnosed as ulcerative colitis (UC) at the first colonoscopy. In the genetic comparison of HLA class I between TA patients with IBD and those without IBD, HLA-B*52:01 and C*12:02 were more frequent in the IBD-TA group (P = 0.001 and P = 0.009, respectively). Meanwhile, HLA-DRB-1*15:02, DQA-1*01:03, DQB-1*06:01, and DPB-1*09:01 as HLA class II were positively associated with IBD-TA (P = 0.004, P = 0.019, P = 0.019, and P = 0.002, respectively). IL12B rs6871626 did not show an association with IBD-TA compared with that with TA without IBD. CONCLUSIONS: The endoscopic findings of IBD-TA at initial diagnosis were atypical for UC or Crohn's disease. IBD-TA possessed the HLA haplotype, which had a susceptible effect on UC.

Comparison of magnetic resonance and balloon enteroscopic examination of the small intestine in patients with Crohn's disease.

BACKGROUND & AIMS: Magnetic resonance (MR) enterography is a recommended imaging technique for detecting intestinal involvement in Crohn's disease (CD). However, the diagnostic accuracy of MR enterography has not been compared directly what that of enteroscopy of the jejunum and proximal ileum. We evaluated the usefulness of MR enterocolonography (MREC) by comparing its findings with those from balloon-assisted enteroscopy. METHODS: In a prospective study, MREC and enteroscopy were performed within 3 days of each other on 100 patients. Ulcerative lesions and all mucosal lesions were evaluated. Physicians and radiologists were blinded to results from other studies. Findings from MREC were compared directly with those from enteroscopy; the sensitivity and specificity with which MREC detected CD lesions were assessed. RESULTS: MREC detected ulcerative lesions and all mucosal lesions in the small intestine with 82.4% sensitivity (95% confidence interval [CI], 75.4%-87.7%) and 67.5% sensitivity (95% CI, 63.1%-70.0%); specificity values were 87.6% (95% CI, 83.7%-90.6%) and 94.8% (95% CI, 90.1%-97.5%). MREC detected major stenosis with 58.8% sensitivity (95% CI, 37.6%-77.2%) and 90.0% specificity (95% CI, 88.4%-91.5%) and all stenoses with 40.8% sensitivity (95% CI, 30.8%-49.4%) and 93.7% specificity (95% CI, 91.1%-95.9%). CONCLUSIONS: MREC is useful for detecting active lesions in the small intestine. However, MR imaging is less sensitive for detecting intestinal damage, such as stenoses. Enteroscopy is preferred for identifying intestinal damage. Suitable imaging approaches should be selected to assess CD lesions in deep small intestine.

Combination of leucine-rich alpha-2 glycoprotein and fecal markers detect Crohn's disease activity confirmed by balloon-assisted enteroscopy.

BACKGROUND/AIMS: Endoscopic activity confirmed by enteroscopy is associated with poor clinical outcome in Crohn's disease (CD). We investigated which of the existing biomarkers best reflects endoscopic activity in CD patients including the small bowel, and whether their combined use can improve accuracy. METHODS: One hundred and four consecutive patients with ileal and ileocolonic type CD who underwent balloon-assisted enteroscopy (BAE) from October 2021 to August 2022 were enrolled, with clinical and laboratory data prospectively collected and analyzed. RESULTS: Hemoglobin, platelet count, C-reactive protein, leucine-rich alpha-2 glycoprotein (LRG), fecal calprotectin, and fecal hemoglobin all showed significant difference in those with ulcers found on BAE. LRG and fecal calprotectin showed the highest areas under the curve (0.841 and 0.853) for detecting ulcers. LRG showed a sensitivity of 78% and specificity of 80% at a cutoff value of 13 µg/mL, whereas fecal calprotectin showed a sensitivity of 91% and specificity of 67% at a cutoff value of 151 µg/g. Dual positivity for LRG and fecal calprotectin, as well as LRG and fecal hemoglobin, both predicted ulcers with an improved specificity of 92% and 100%. A positive LRG or fecal calprotectin/hemoglobin showed an improved sensitivity of 96% and 91%. Positivity for LRG and either of the fecal biomarkers was associated with increased risk of hospitalization, surgery, and relapse. CONCLUSIONS: The biomarkers LRG, fecal calprotectin, and fecal hemoglobin can serve as noninvasive and accurate tools for assessing activity in CD patients confirmed by BAE, especially when used in combination.