A Systematic Review of Artificial Intelligence and Machine Learning Applications to Inflammatory Bowel Disease, with Practical Guidelines for Interpretation

Artificial Intelligence for Classification of Endoscopic Severity of Inflammatory Bowel Disease: A Systematic Review and Critical Appraisal

BACKGROUND

Endoscopic scoring indices for ulcerative colitis and Crohn's disease are subject to inter-endoscopist variability. There is increasing interest in the development of deep learning models to standardize endoscopic assessment of intestinal diseases. Here, we summarize and critically appraise the literature on artificial intelligence-assisted endoscopic characterization of inflammatory bowel disease severity.

METHODS

A systematic search of Ovid MEDLINE, EMBASE, Cochrane Central Register of Controlled Trials, and IEEE Xplore was performed to identify reports of AI systems used for endoscopic severity classification of IBD. Selected studies were critically appraised for methodological and reporting quality using APPRAISE-AI.

RESULTS

Thirty-one studies published between 2019 and 2024 were included. Of 31 studies, 28 studies examined endoscopic classification of ulcerative colitis and 3 examined Crohn's disease. Researchers sought to accomplish a wide range of classification tasks, including binary and multilevel classification, based on still images or full-length colonoscopy videos. Overall scores for study quality ranged from 41 (moderate quality) to 64 (high quality) out of 100, with 28 out of 31 studies within the moderate quality range. The highest-scoring domains were clinical relevance and reporting quality, while the lowest-scoring domains were robustness of results and reproducibility.

CONCLUSIONS

Multiple AI models have demonstrated the potential for clinical translation for ulcerative colitis. Research concerning the endoscopic severity assessment of Crohn's disease is limited and should be further explored. More rigorous external validation of AI models and increased transparency of data and codes are needed to improve the quality of AI studies.

Automated inflammatory bowel disease detection using wearable bowel sound event spotting

Introduction

Inflammatory bowel disorders may result in abnormal Bowel Sound (BS) characteristics during auscultation. We employ pattern spotting to detect rare bowel BS events in continuous abdominal recordings using a smart T-shirt with embedded miniaturised microphones. Subsequently, we investigate the clinical relevance of BS spotting in a classification task to distinguish patients diagnosed with inflammatory bowel disease (IBD) and healthy controls.

Methods

Abdominal recordings were obtained from 24 patients with IBD with varying disease activity and 21 healthy controls across different digestive phases. In total, approximately 281 h of audio data were inspected by expert raters and thereof 136 h were manually annotated for BS events. A deep-learning-based audio pattern spotting algorithm was trained to retrieve BS events. Subsequently, features were extracted around detected BS events and a Gradient Boosting Classifier was trained to classify patients with IBD vs. healthy controls. We further explored classification window size, feature relevance, and the link between BS-based IBD classification performance and IBD activity.

Results

Stratified group K-fold cross-validation experiments yielded a mean area under the receiver operating characteristic curve ≥0.83 regardless of whether BS were manually annotated or detected by the BS spotting algorithm.

Discussion

Automated BS retrieval and our BS event classification approach have the potential to support diagnosis and treatment of patients with IBD.

Artificial intelligence to revolutionize IBD clinical trials: a comprehensive review

Integrating artificial intelligence (AI) into clinical trials for inflammatory bowel disease (IBD) has potential to be transformative to the field. This article explores how AI-driven technologies, including machine learning (ML), natural language processing, and predictive analytics, have the potential to enhance important aspects of IBD trials-from patient recruitment and trial design to data analysis and personalized treatment strategies. As AI advances, it has potential to improve long-standing challenges in trial efficiency, accuracy, and personalization with the goal of accelerating the discovery of novel therapies and improve outcomes for people living with IBD. AI can streamline multiple trial phases, from target identification and patient recruitment to data analysis and monitoring. By integrating multi-omics data, electronic health records, and imaging repositories, AI can uncover molecular targets and personalize trial strategies, ultimately expediting drug development. However, the adoption of AI in IBD clinical trials encounters significant challenges. These include technical barriers in data integration, ethical concerns regarding patient privacy, and regulatory issues related to AI validation standards. Additionally, AI models risk producing biased outcomes if training datasets lack diversity, potentially impacting underrepresented populations in clinical trials. Addressing these limitations requires standardized data formats, interdisciplinary collaboration, and robust ethical frameworks to ensure inclusivity and accuracy. Continued partnerships among clinicians, researchers, data scientists, and regulators will be essential to establish transparent, patient-centered AI frameworks. By overcoming these obstacles, AI has the potential to enhance the efficiency, equity, and efficacy of IBD clinical trials, ultimately benefiting patient care.

Ulcerative Colitis Severity Classification and Localized Extent (UC-SCALE): An Artificial Intelligence Scoring System for a Spatial Assessment of Disease Severity in Ulcerative Colitis

BACKGROUND AND AIMS

Validated scoring methods such as the Mayo Clinic Endoscopic Subscore (MCES) evaluate ulcerative colitis (UC) severity at the worst colon segment, without considering disease extent. We present the Ulcerative Colitis Severity Classification and Localized Extent (UC-SCALE) algorithm, which provides a comprehensive and automated evaluation of endoscopic severity and disease extent in UC.

METHODS

Ulcerative Colitis Severity Classification and Localized Extent consists of 3 main elements: (1) a quality filter selecting readable images (frames) from colonoscopy videos, (2) a scoring system assigning an MCES to each readable frame, and (3) a camera localization algorithm assigning each frame to a location within the colon. Ulcerative Colitis Severity Classification and Localized Extent was trained and tested using 4326 sigmoidoscopy videos from phase III Etrolizumab clinical trials.

RESULTS

The high agreement between UC-SCALE and central reading at the level of the colon section (𝜅 = 0.80), and the agreement between central and local reading (𝜅 = 0.84), suggested a similar inter-rater agreement between UC-SCALE and experienced readers. Furthermore, UC-SCALE correlated with disease activity markers such calprotectin, C-reactive protein and patient-reported outcomes, Physician Global Assessment and Geboes Histologic scores (rs 0.40-0.55, ps < 0.0001). Finally, the value of using UC-SCALE was demonstrated by assessing individual endoscopic severity between baseline and induction.

CONCLUSIONS

Our fully automated scoring system enables accurate, objective, and localized assessment of endoscopic severity in UC patients. In addition, we provide a topological representation of the score as a marker of disease severity that correlates highly with clinical metrics. Ulcerative Colitis Severity Classification and Localized Extent reproduces central reading and holds promise to enhance disease severity evaluation in both clinical trials and everyday practice.

The global research of artificial intelligence on inflammatory bowel disease: A bibliometric analysis

Aims

This study aimed to evaluate the related research on artificial intelligence (AI) in inflammatory bowel disease (IBD) through bibliometrics analysis and identified the research basis, current hotspots, and future development.

Methods

The related literature was acquired from the Web of Science Core Collection (WoSCC) on 31 December 2024. Co-occurrence and cooperation relationship analysis of (cited) authors, institutions, countries, cited journals, references, and keywords in the literature were carried out through CiteSpace 6.1.R6 software and the Online Analysis platform of Literature Metrology. Meanwhile, relevant knowledge maps were drawn, and keywords clustering analysis was performed.

Results

According to WoSCC, 1919 authors, 790 research institutions, 184 journals, and 49 countries/regions published 176 AI-related papers in IBD during 1999-2024. The number of papers published has increased significantly since 2019, reaching a maximum by 2023. The United States had the highest number of publications and the closest collaboration with other countries. The clustering analysis showed that the earliest studies focused on "psychometric value" and then moved to "deep learning model," "intestinal ultrasound," and "new diagnostic strategies."

Conclusion

This study is the first bibliometric analysis to summarize the current status and to visually reveal the development trends and future research hotspots of the application of AI in IBD. The application of AI in IBD is still in its infancy, and the focus of this field will shift to improving the efficiency of diagnosis and treatment through deep learning techniques, big data-based treatment, and prognosis prediction.

Effectiveness of Telemedicine in Inflammatory Bowel Disease in Russia: TIGE-Rus (Telemonitoring for IBD Goodness Examination in Russia) Study Protocol of a Randomized Controlled Trial

Background: Inflammatory bowel diseases (IBD), associated with a significant burden on patients' lives, are becoming increasingly common. Patients with IBD need continuous treatment and lifelong monitoring, which could be achieved by telemonitoring. Telemonitoring has been shown to be effective in improving outcomes for patients with IBD, and can provide a more convenient and accessible way for patients to receive care. However, the certainty of evidence remains low. This article outlines the methodology of a randomized control study that aims to assess the efficacy of telemonitoring compared to face-to-face follow-up for patients with IBD in Russia, hypothesizing that the implementation of telemonitoring will lead to improvement in clinical, social, and organizational areas. Methods: The TIGE-Rus study is a randomized controlled trial. The study consists of three stages, including selection of patients and random assignment into two groups with a ratio of 1:1, follow-up care using telemonitoring or face-to-face appointments, and evaluation and comparison of follow-up efficacy in both groups. In the first stage, all patients will undergo laboratory tests and instrumental examinations, and fill out questionnaires to measure disease activity, quality of life, medication adherence, psychological well-being, and satisfaction with medical care. In the second stage, the control group will receive standard care while the telemonitoring group will have access to a web platform where they can report their clinical activity, fill out questionnaires, and have online consultations with gastroenterologists. The gastroenterologists will also make monthly phone calls to each patient in the telemonitoring group to monitor their progress. In the third stage of the study, both the telemonitoring group and the control group will be re-hospitalized after six months of monitoring. IBD activity will be evaluated through laboratory and instrumental examinations. Additionally, all the participants will complete questionnaires to assess the disease activity, medication adherence, quality of life, psychological well-being, and satisfaction with medical care in both groups. Conclusions: The trial will explore whether telemonitoring is effective in improving clinical, social, and organizational aspects in the management of patients with IBD in the setting of the Russian healthcare system.

Wise Roles and Future Visionary Endeavors of Current Emperor: Advancing Dynamic Methods for Longitudinal Microbiome Meta-Omics Data in Personalized and Precision Medicine

Understanding the etiological complexity of diseases requires identifying biomarkers longitudinally associated with specific phenotypes. Advanced sequencing tools generate dynamic microbiome data, providing insights into microbial community functions and their impact on health. This review aims to explore the current roles and future visionary endeavors of dynamic methods for integrating longitudinal microbiome multi-omics data in personalized and precision medicine. This work seeks to synthesize existing research, propose best practices, and highlight innovative techniques. The development and application of advanced dynamic methods, including the unified analytical frameworks and deep learning tools in artificial intelligence, are critically examined. Aggregating data on microbes, metabolites, genes, and other entities offers profound insights into the interactions among microorganisms, host physiology, and external stimuli. Despite progress, the absence of gold standards for validating analytical protocols and data resources of various longitudinal multi-omics studies remains a significant challenge. The interdependence of workflow steps critically affects overall outcomes. This work provides a comprehensive roadmap for best practices, addressing current challenges with advanced dynamic methods. The review underscores the biological effects of clinical, experimental, and analytical protocol settings on outcomes. Establishing consensus on dynamic microbiome inter-studies and advancing reliable analytical protocols are pivotal for the future of personalized and precision medicine.

EUS in children with eosinophilic oesophagitis - a new method of measuring oesophageal total wall thickness area. An artificial intelligence application feasibility study. A pilot study

Aim

In the study, we aimed to introduce a formula for measuring the oesophageal total wall thickness area, which could be used for developing an artificial intelligence-based algorithm for the detection of patients whose total wall thickness area exceeds the norms.

Material and methods

Mathematical formulas for measuring the square area of the oesophageal total wall thickness area were introduced and applied. Children were grouped according to their weight in clusters. For each cluster, the range (minimal and maximal value) were established. The measurements were done by using the formula for the area of the circular ring according to the formula A = n (B2-b2); the product of n and subtraction square b (smaller radius) and square B (bigger radius). The basic data for our calculations were derived from papers published by Dalby et al., 2010 and Loff et al., 2022.

Results

The square area (in mm2) of the oesophageal wall was calculated and proposed to be introduced for further analysis. This value set could be used for creating an algorithm for computer-aided analysis of patients diagnosed with sonographic examination and isolating patients for surveillance. Our newly introduced approach could be implemented in sonographic, computer tomography, and magnetic resonance examinations in eosinophilic oesophagitis and other oesophageal diseases.

Conclusions

Total wall thickness area could be used for monitoring children with eosinophilic oesophagitis and other oesophageal diseases. The method could also be applied for adults. Therefore, it can be a foundation for further progress with applying artificial intelligence algorithms.

Developing Mobile Health Applications for Inflammatory Bowel Disease: A Systematic Review of Features and Technologies

Background

Patients with inflammatory bowel disease (IBD) require lifelong treatment, which significantly impacts their quality of life. Self-management of this disease is an effective factor in managing chronic conditions and improving patients' quality of life. The use of mobile applications is a novel approach to providing self-management models and healthcare services for patients with IBD. The present systematic review aimed to identify the features and technologies used in the development of IBD disease management applications.

Methods

This systematic review was conducted according to PRISMA guidelines in PubMed, Scopus, and Web of Sciences databases up to August 8, 2023, which included initial searches, screening studies, assessing eligibility and risk of bias, and study selection. The data extraction form was based on the study objectives, including bibliographic information from articles, such as the first author's name, year of publication, country of origin, and details related to mobile health applications, such as the name of the application, features and technologies used, advantages and disadvantages, main outcomes, and other results. The content of the research was analyzed according to the research objectives.

Results

In the initial review of four databases, a total of 160 articles were retrieved and subsequently entered into EndNote. After removing duplicates and irrelevant studies based on title, abstract, and full-text assessments, 12 articles were finally selected. The studies were conducted between the years 2015 and 2024. 100% of the applications designed for patients with IBD were aimed at treatment, 83% were for self-management of the disease, and 33% of the applications were intended for disease diagnosis. The features of IBD management applications were categorized into four groups: education, monitoring, counseling, and diagnosis and treatment.

Conclusion

Various mobile applications have been developed for the management of IBD, each differing in features and technologies used. While current IBD applications have limited capabilities in diagnosing disease severity, they still hold significant potential in empowering patients through education, counseling, and monitoring. The integration of artificial intelligence and decision support systems may enhance the effectiveness and reliability of these applications.

Defining management strategies for acute severe ulcerative colitis using predictive models: a simulation-modeling study

BACKGROUND/AIMS

Robust management algorithms are required to reduce the residual risk of colectomy in acute severe ulcerative colitis (ASUC) refractory to standard infliximab salvage therapy. The aim of this study was to evaluate the performance and benefits of alternative ASUC management strategies using simulated prediction models of varying accuracy.

METHODS

This was a simulation-based modeling study using a hypothetical cohort of 5,000 steroid-refractory ASUC patients receiving standard infliximab induction. Simulated predictive models were used to risk-stratify patients and escalate treatment in patients at high risk of failing standard infliximab induction. The main outcome of interest was colectomy by 3 months.

RESULTS

The 3-month colectomy rate in the base scenario where all 5,000 patients received standard infliximab induction was 23%. The best-performing management strategy assigned high-risk patients to sequential Janus kinase inhibitor inhibition and mediumrisk patients to accelerated infliximab induction. Using a 90% area under the curve (AUC) prediction model and optimistic treatment efficacy assumptions, this strategy reduced the 3-month colectomy rate to 8% (65% residual risk reduction). Using an 80% AUC prediction model with only modest treatment efficacy assumptions, the 3-month colectomy rate was reduced to 15% (35% residual risk reduction). Overall management strategy efficacy was highly dependent on predictive model accuracy and underlying treatment efficacy assumptions.

CONCLUSIONS

This is the first study to simulate predictive model-based management strategies in steroid-refractory ASUC and evaluate their effect on short-term colectomy rates. Future studies on predictive model development should incorporate simulation studies to better understand their expected benefit.

Machine learning-derived phenotypic trajectories of asthma and allergy in children and adolescents: protocol for a systematic review

INTRODUCTION

Development of asthma and allergies in childhood/adolescence commonly follows a sequential progression termed the 'atopic march'. Recent reports indicate, however, that these diseases are composed of multiple distinct phenotypes, with possibly differential trajectories. We aim to synthesise the current literature in the field of machine learning-based trajectory studies of asthma/allergies in children and adolescents, summarising the frequency, characteristics and associated risk factors and outcomes of identified trajectories and indicating potential directions for subsequent research in replicability, pathophysiology, risk stratification and personalised management. Furthermore, methodological approaches and quality will be critically appraised, highlighting trends, limitations and future perspectives.

METHODS AND ANALYSES

10 databases (CAB Direct, CINAHL, Embase, Google Scholar, PsycInfo, PubMed, Scopus, Web of Science, WHO Global Index Medicus and WorldCat Dissertations and Theses) will be searched for observational studies (including conference abstracts and grey literature) from the last 10 years (2013-2023) without restriction by language. Screening, data extraction and assessment of quality and risk of bias (using a custom-developed tool) will be performed independently in pairs. The characteristics of the derived trajectories will be narratively synthesised, tabulated and visualised in figures. Risk factors and outcomes associated with the trajectories will be summarised and pooled estimates from comparable numerical data produced through random-effects meta-analysis. Methodological approaches will be narratively synthesised and presented in tabulated form and figure to visualise trends.

ETHICS AND DISSEMINATION

Ethical approval is not warranted as no patient-level data will be used. The findings will be published in an international peer-reviewed journal.

PROSPERO REGISTRATION NUMBER

CRD42023441691.

Histopathology imaging and clinical data including remission status in pediatric inflammatory bowel disease

The incidence of inflammatory bowel disease (IBD) is increasing annually. Children with IBD often suffer significant morbidity due to physical and emotional effects of the disease and treatment. Corticosteroids, often a component of therapy, carry undesirable side effects with long term use. Steroid-free remission has become a standard for care-quality improvement. Anticipating therapeutic outcomes is difficult, with treatments often leveraged in a trial-and-error fashion. Artificial intelligence (AI) has demonstrated success in medical imaging classification tasks. Predicting patients who will attain remission will help inform treatment decisions. The provided dataset comprises 951 tissue section scans (167 whole-slides) obtained from 18 pediatric IBD patients. Patient level structured data include IBD diagnosis, 12- and 52-week steroid use and name, and remission status. Each slide is labelled with biopsy site and normal or abnormal classification per the surgical pathology report. Each tissue section scan from an abnormal slide is further classified by an experienced pathologist. Researchers utilizing this dataset may select from the provided outcomes or add labels and annotations from their own institutions.

Can artificial intelligence replace endoscopists when assessing mucosal healing in ulcerative colitis? A systematic review and diagnostic test accuracy meta-analysis

BACKGROUNDS AND AIMS

Mucosal healing (MH) in inflammatory bowel diseases (IBD) is an important landmark for clinical decision making. Artificial intelligence systems (AI) that automatically deliver the grade of endoscopic inflammation may solve moderate interobserver agreement and the need of central reading in clinical trials.

METHODS

We performed a systematic review of EMBASE and MEDLINE databases up to 01/12/2022 following PRISMA and the Joanna Briggs Institute methodologies to answer the following question: "Can AI replace endoscopists when assessing MH in IBD?". The research was restricted to ulcerative colitis (UC), and a diagnostic odds ratio (DOR) meta-analysis was performed. Risk of bias was evaluated with QUADAS-2 tool.

RESULTS

A total of 21 / 739 records were selected for full text evaluation, and 12 were included in the meta-analysis. Deep learning algorithms based on convolutional neural networks architecture achieved a satisfactory performance in evaluating MH on UC, with sensitivity, specificity, DOR and SROC of respectively 0.91(CI95 %:0.86-0.95);0.89(CI95 %:0.84-0.93);92.42(CI95 %:54.22-157.53) and 0.957 when evaluating fixed images (n = 8) and 0.86(CI95 %:0.75-0.93);0.91(CI95 %:0.87-0.94);70.86(CI95 %:24.63-203.86) and 0.941 when evaluating videos (n = 6). Moderate-high levels of heterogeneity were noted, limiting the quality of the evidence.

CONCLUSIONS

AI systems showed high potential in detecting MH in UC with optimal diagnostic performance, although moderate-high heterogeneity of the data was noted. Standardised and shared AI training may reduce heterogeneity between systems.

Advances in Inflammatory Bowel Disease Diagnostics: Machine Learning and Genomic Profiling Reveal Key Biomarkers for Early Detection

This study, utilizing high-throughput technologies and Machine Learning (ML), has identified gene biomarkers and molecular signatures in Inflammatory Bowel Disease (IBD). We could identify significant upregulated or downregulated genes in IBD patients by comparing gene expression levels in colonic specimens from 172 IBD patients and 22 healthy individuals using the GSE75214 microarray dataset. Our ML techniques and feature selection methods revealed six Differentially Expressed Gene (DEG) biomarkers (VWF, IL1RL1, DENND2B, MMP14, NAAA, and PANK1) with strong diagnostic potential for IBD. The Random Forest (RF) model demonstrated exceptional performance, with accuracy, F1-score, and AUC values exceeding 0.98. Our findings were rigorously validated with independent datasets (GSE36807 and GSE10616), further bolstering their credibility and showing favorable performance metrics (accuracy: 0.841, F1-score: 0.734, AUC: 0.887). Our functional annotation and pathway enrichment analysis provided insights into crucial pathways associated with these dysregulated genes. DENND2B and PANK1 were identified as novel IBD biomarkers, advancing our understanding of the disease. The validation in independent cohorts enhances the reliability of these findings and underscores their potential for early detection and personalized treatment of IBD. Further exploration of these genes is necessary to fully comprehend their roles in IBD pathogenesis and develop improved diagnostic tools and therapies. This study significantly contributes to IBD research with valuable insights, potentially greatly enhancing patient care.

Establishing a machine learning model based on dual-energy CT enterography to evaluate Crohn's disease activity

OBJECTIVES

The simplified endoscopic score of Crohn's disease (SES-CD) is the gold standard for quantitatively evaluating Crohn's disease (CD) activity but is invasive. This study aimed to develop and validate a machine learning (ML) model based on dual-energy CT enterography (DECTE) to noninvasively evaluate CD activity.

METHODS

We evaluated the activity in 202 bowel segments of 46 CD patients according to the SES-CD score and divided the segments randomly into training set and testing set at a ratio of 7:3. Least absolute shrinkage and selection operator (LASSO) was used for feature selection, and three models based on significant parameters were established based on logistic regression. Model performance was evaluated using receiver operating characteristic (ROC), calibration, and clinical decision curves.

RESULTS

There were 110 active and 92 inactive bowel segments. In univariate analysis, the slope of spectral curve in the venous phases (λHU-V) has the best diagnostic performance, with an area under the ROC curve (AUC) of 0.81 and an optimal threshold of 1.975. In the testing set, the AUC of the three models established by the 7 variables to differentiate CD activity was 0.81-0.87 (DeLong test p value was 0.071-0.766, p > 0.05), and the combined model had the highest AUC of 0.87 (95% confidence interval (CI): 0.779-0.959).

CONCLUSIONS

The ML model based the DECTE can feasibly evaluate CD activity, and DECTE parameters provide a quantitative analysis basis for evaluating specific bowel activities in CD patients.

CRITICAL RELEVANCE STATEMENT

The machine learning model based on dual-energy computed tomography enterography can be used for evaluating Crohn's disease activity noninvasively and quantitatively.

KEY POINTS

Dual-energy CT parameters are related to Crohn's disease activity. Three machine learning models effectively evaluated Crohn's disease activity. Combined models based on conventional and dual-energy CT have the best performance.

Imaging and endoscopic tools in pediatric inflammatory bowel disease: What's new?

Pediatric inflammatory bowel disease (IBD) is a chronic inflammatory disorder, with increasing incidence and prevalence worldwide. There have been recent advances in imaging and endoscopic technology for disease diagnosis, treatment, and monitoring. Intestinal ultrasound, including transabdominal, transperineal, and endoscopic, has been emerging for the assessment of transmural bowel inflammation and disease complications (e.g., fistula, abscess). Aside from surgery, IBD-related intestinal strictures now have endoscopic treatment options including through-the-scope balloon dilatation, injection, and needle knife stricturotomy and new evaluation tools such as endoscopic functional lumen imaging probe. Unsedated transnasal endoscopy may have a role in patients with upper gastrointestinal Crohn's disease or those with IBD with new upper gastrointestinal symptoms. Improvements to dysplasia screening in pediatric patients with longstanding colonic disease or primary sclerosing cholangitis hold promise with the addition of virtual chromoendoscopy and ongoing research in the field of artificial intelligence-assisted endoscopic detection. Artificial intelligence and machine learning is a rapidly evolving field, with goals of further personalizing IBD diagnosis and treatment selection as well as prognostication. This review summarized these advancements, focusing on pediatric patients with IBD.

Supervised Machine Learning Classifies Inflammatory Bowel Disease Patients by Subtype Using Whole Exome Sequencing Data

BACKGROUND

Inflammatory bowel disease [IBD] is a chronic inflammatory disorder with two main subtypes: Crohn's disease [CD] and ulcerative colitis [UC]. Prompt subtype diagnosis enables the correct treatment to be administered. Using genomic data, we aimed to assess machine learning [ML] to classify patients according to IBD subtype.

METHODS

Whole exome sequencing [WES] from paediatric/adult IBD patients was processed using an in-house bioinformatics pipeline. These data were condensed into the per-gene, per-individual genomic burden score, GenePy. Data were split into training and testing datasets [80/20]. Feature selection with a linear support vector classifier, and hyperparameter tuning with Bayesian Optimisation, were performed [training data]. The supervised ML method random forest was utilised to classify patients as CD or UC, using three panels: 1] all available genes; 2] autoimmune genes; 3] 'IBD' genes. ML results were assessed using area under the receiver operating characteristics curve [AUROC], sensitivity, and specificity on the testing dataset.

RESULTS

A total of 906 patients were included in analysis [600 CD, 306 UC]. Training data included 488 patients, balanced according to the minority class of UC. The autoimmune gene panel generated the best performing ML model [AUROC = 0.68], outperforming an IBD gene panel [AUROC = 0.61]. NOD2 was the top gene for discriminating CD and UC, regardless of the gene panel used. Lack of variation in genes with high GenePy scores in CD patients was the best classifier of a diagnosis of UC.

DISCUSSION

We demonstrate promising classification of patients by subtype using random forest and WES data. Focusing on specific subgroups of patients, with larger datasets, may result in better classification.

Artificial Intelligence in Inflammatory Bowel Disease Endoscopy: Implications for Clinical Trials

Artificial intelligence shows promise for clinical research in inflammatory bowel disease endoscopy. Accurate assessment of endoscopic activity is important in clinical practice and inflammatory bowel disease clinical trials. Emerging artificial intelligence technologies can increase efficiency and accuracy of assessing the baseline endoscopic appearance in patients with inflammatory bowel disease and the impact that therapeutic interventions may have on mucosal healing in both of these contexts. In this review, state-of-the-art endoscopic assessment of mucosal disease activity in inflammatory bowel disease clinical trials is described, covering the potential for artificial intelligence to transform the current paradigm, its limitations, and suggested next steps. Site-based artificial intelligence quality evaluation and inclusion of patients in clinical trials without the need for a central reader is proposed; for following patient progress, a second reading using AI alongside a central reader with expedited reading is proposed. Artificial intelligence will support precision endoscopy in inflammatory bowel disease and is on the threshold of advancing inflammatory bowel disease clinical trial recruitment.

Diagnostic accuracy of convolutional neural network-based machine learning algorithms in endoscopic severity prediction of ulcerative colitis: a systematic review and meta-analysis

BACKGROUND AND AIMS

Endoscopic assessment of ulcerative colitis (UC) can be performed by using the Mayo Endoscopic Score (MES) or the Ulcerative Colitis Endoscopic Index of Severity (UCEIS). In this meta-analysis, we assessed the pooled diagnostic accuracy parameters of deep machine learning by means of convolutional neural network (CNN) algorithms in predicting UC severity on endoscopic images.

METHODS

Databases including MEDLINE, Scopus, and Embase were searched in June 2022. Outcomes of interest were the pooled accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Standard meta-analysis methods used the random-effects model, and heterogeneity was assessed using the I2statistics.

RESULTS

Twelve studies were included in the final analysis. The pooled diagnostic parameters of CNN-based machine learning algorithms in endoscopic severity assessment of UC were as follows: accuracy 91.5% (95% confidence interval [CI], 88.3-93.8; I2 = 84%), sensitivity 82.8% (95% CI, 78.3-86.5; I2 = 89%), specificity 92.4% (95% CI, 89.4-94.6; I2 = 84%), PPV 86.6% (95% CI, 82.3-90; I2 = 89%), and NPV 88.6% (95% CI, 85.7-91; I2 = 78%). Subgroup analysis revealed significantly better sensitivity and PPV with the UCEIS scoring system compared with the MES (93.6% [95% CI, 87.5-96.8; I2 = 77%] vs 82% [95% CI, 75.6-87; I2 = 89%], P = .003, and 93.6% [95% CI, 88.7-96.4; I2 = 68%] vs 83.6% [95% CI, 76.8-88.8; I2 = 77%], P = .007, respectively).

CONCLUSIONS

CNN-based machine learning algorithms demonstrated excellent pooled diagnostic accuracy parameters in the endoscopic severity assessment of UC. Using UCEIS scores in CNN training might offer better results than the MES. Further studies are warranted to establish these findings in real clinical settings.

Artificial-intelligence-based decision support tools for the differential diagnosis of colitis

BACKGROUND

Whereas Artificial Intelligence (AI) based tools have recently been introduced in the field of gastroenterology, application in inflammatory bowel disease (IBD) is in its infancies. We established AI-based algorithms to distinguish IBD from infectious and ischemic colitis using endoscopic images and clinical data.

METHODS

First, we trained and tested a Convolutional Neural Network (CNN) using 1796 real-world images from 494 patients, presenting with three diseases (IBD [n = 212], ischemic colitis [n = 157], and infectious colitis [n = 125]). Moreover, we evaluated a Gradient Boosted Decision Trees (GBDT) algorithm using five clinical parameters as well as a hybrid approach (CNN + GBDT). Patients and images were randomly split into two completely independent datasets. The proposed approaches were benchmarked against each other and three expert endoscopists on the test set.

RESULTS

For the image-based CNN, the GBDT algorithm and the hybrid approach global accuracies were .709, .792, and .766, respectively. Positive predictive values were .602, .702, and .657. Global areas under the receiver operating characteristics (ROC) and precision recall (PR) curves were .727/.585, .888/.823, and .838/.733, respectively. Global accuracy did not differ between CNN and endoscopists (.721), but the clinical parameter-based GBDT algorithm outperformed CNN and expert image classification.

CONCLUSIONS

Decision support systems exclusively based on endoscopic image analysis for the differential diagnosis of colitis, representing a complex clinical challenge, seem not yet to be ready for primetime and more diverse image datasets may be necessary to improve performance in future development. The clinical value of the proposed clinical parameters algorithm should be evaluated in prospective cohorts.

Performance of Machine Learning Algorithms for Predicting Disease Activity in Inflammatory Bowel Disease

This study aimed to explore the effectiveness of predicting disease activity in patients with inflammatory bowel disease (IBD), using machine learning (ML) models. A retrospective research was undertaken on IBD patients who were admitted into the First Affiliated Hospital of Wenzhou Medical University between September 2011 and September 2019. At first, data were randomly split into a 3:1 ratio of training to test set. The least absolute shrinkage and selection operator (LASSO) algorithm was applied to reduce the dimension of variables. These variables were used to generate seven ML algorithms, namely random forests (RFs), adaptive boosting (AdaBoost), K-nearest neighbors (KNNs), support vector machines (SVMs), naïve Bayes (NB), ridge regression, and eXtreme gradient boosting (XGBoost) to train to predict disease activity in IBD patients. SHapley Additive exPlanation (SHAP) analysis was performed to rank variable importance. A total of 876 participants with IBD, consisting of 275 ulcerative colitis (UC) and 601 Crohn's disease (CD), were retrospectively enrolled in the study. Thirty-three variables were obtained from the clinical characteristics and laboratory tests of the participants. Finally, after LASSO analysis, 11 and 5 variables were screened out to construct ML models for CD and UC, respectively. All seven ML models performed well in predicting disease activity in the CD and UC test sets. Among these ML models, SVM was more effective in predicting disease activity in the CD group, whose AUC reached 0.975, sensitivity 0.947, specificity 0.920, and accuracy 0.933. AdaBoost performed best for the UC group, with an AUC of 0.911, sensitivity 0.844, specificity 0.875, and accuracy 0.855. ML algorithms were available and capable of predicting disease activity in IBD patients. Based on clinical and laboratory variables, ML algorithms demonstrate great promise in guiding physicians' decision-making.

Prediction of Crohn's Disease Stricturing Phenotype Using a NOD2-derived Genomic Biomarker

BACKGROUND

Crohn's disease (CD) is highly heterogenous and may be complicated by stricturing behavior. Personalized prediction of stricturing will inform management. We aimed to create a stricturing risk stratification model using genomic/clinical data.

METHODS

Exome sequencing was performed on CD patients, and phenotype data retrieved. Biallelic variants in NOD2 were identified. NOD2 was converted into a per-patient deleteriousness metric ("GenePy"). Using training data, patients were stratified into risk groups for fibrotic stricturing using NOD2. Findings were validated in a testing data set. Models were modified to include disease location at diagnosis. Cox proportional hazards assessed performance.

RESULTS

Six hundred forty-five patients were included (373 children and 272 adults); 48 patients fulfilled criteria for monogenic NOD2-related disease (7.4%), 24 of whom had strictures. NOD2 GenePy scores stratified patients in training data into 2 risk groups. Within testing data, 30 of 161 patients (18.6%) were classified as high-risk based on the NOD2 biomarker, with stricturing in 17 of 30 (56.7%). In the low-risk group, 28 of 131 (21.4%) had stricturing behavior. Cox proportional hazards using the NOD2 risk groups demonstrated a hazard ratio (HR) of 2.092 (P = 2.4 × 10-5), between risk groups. Limiting analysis to patients diagnosed aged < 18-years improved performance (HR-3.164, P = 1 × 10-6). Models were modified to include disease location, such as terminal ileal (TI) disease or not. Inclusion of NOD2 risk groups added significant additional utility to prediction models. High-risk group pediatric patients presenting with TI disease had a HR of 4.89 (P = 2.3 × 10-5) compared with the low-risk group patients without TI disease.

CONCLUSIONS

A NOD2 genomic biomarker predicts stricturing risk, with prognostic power improved in pediatric-onset CD. Implementation into a clinical setting can help personalize management.