The Emergence and Significance of Using Computer Imaging for Measuring Inflammatory Bowel Disease (IBD) Activity

Inflammatory bowel disease (IBD), an umbrella term that includes chronic immune-mediated disorders like Crohn’s disease (CD) and ulcerative colitis (UC), affects more than 6 million people worldwide.[1] IBD can present differently in different people, making the diagnosis somewhat challenging. There are many diagnostic criteria that clinicians use to identify IBD including symptoms, biomarkers, endoscopy and histology. In the end, the diagnosis will ultimately depend on the judgement of the physician, which can affect the reproducibility of disease severity assessments and limit diagnosis for patients who do not have ready access to IBD experts.[2] For patients suffering from IBD, a delay in diagnosis can lead to less favorable response to therapies, an increased rate of complications, and potentially increase the need for surgical intervention.[3,4]

Increasingly, healthcare providers are looking to quantify diagnosis. In order to do so, they are turning to the use of computer vision (CV) methods and artificial intelligence (AI) to better determine disease activity and severity in patients to establish treatment approaches.[5]

Advantages of Computer Imaging in Measuring Disease Activity and Severity

CV is the application of AI in the visual inspection of medical imaging. Historically, imaging experts have been required to analyze images to classify and make distinctions between disease severity grades in order to diagnose a patient, which can require extensive high-quality video review and scoring.[6] When applied to IBD endoscopic imaging, CV may offer high accuracy, precision, and reproducibility, factors which can be  time-consuming for imaging experts.[5] These automated systems can be trained to interpret imaging, using large sets of endoscopic videos collected from IBD patients in clinical trials as well as perform detection and characterization of mucosal abnormalities, classify the severity of inflammation, and assess mucosal healing.[5] Solutions such as these have the potential of reducing the variation and time to diagnosis, which ultimately could improve the quality of care for IBD patients.[5]

Challenges in Use of Computer Imaging

Computer vision and AI for medical imaging analysis are in their infancy and come with limitations and challenges in IBD assessment.[5] Many CV-based applications rely on recognizing patterns and identifying key features based on training from specific data sets, so the variability in assessments by physicians in their clinics can lead to challenges in generalizability of the algorithms across healthcare settings.[5] Additionally, CV solutions do not speak to causal inference or functionality, and are ultimately only as useful as the available data – if there is missing information or specific clinical context for a patient, it can limit the applicability of CV.[5]

Additionally, many CV solutions limit their interpretability and provide only probabilities of a prediction, not an absolute answer.[5]

Ongoing Work to Advance Treatment for Patients in Need

We are dedicated to developing solutions for patients in need, and are always looking for the latest technologies to help in that quest. The drive for new solutions for patients propels us at Janssen, as we continue to advance the latest science in immune-mediated diseases. Our efforts include a collaboration with the University of Michigan and Dr. Ryan Stidham on further research on automatic estimation of disease severity from endoscopic videos using multi-instance learning.[7]

Janssen is pleased to be presenting new data in the use of CV and AI for measuring IBD activity and severity at Digestive Disease Week (DDW) 2023, taking place in Chicago, Illinois and virtually May 6-9.

To learn more about our work and our presence at this year’s DDW meeting, please visit: https://bit.ly/3RFltMn.

One of the most effective ways of understanding and improving health outcomes for patients in need is by having a workforce that understands the critical needs of all patients. I invite you to learn more about how to be a part of our efforts with careers in Immunology at Janssen by visiting us at: https://bit.ly/3hQ8Q24.

#DDW2023 #IBD #CrohnsDisease #Ulcerativecolitis #Gastroenterology #Immunology #MyCompany

References:

1. GBD 2017 Inflammatory Bowel Disease Collaborators. The Global, Regional, and National Burden of Inflammatory Bowel Disease in 195 Countries and Territories, 1990-2017: A Systematic Analysis for the Global Burden of Disease Study 2017. The Lancet Gastroenterology & hepatology, 5(1), 17–30. https://doi.org/10.1016/S2468-1253(19)30333-4
2. Chen, D. et al., Application of Artificial Intelligence to Clinical Practice in Inflammatory Bowel Disease – What the Clinician Needs to Know, Journal of Crohn's and Colitis, Volume 16, Issue 3, March 2022, Pages 460–471, https://doi.org/10.1093/ecco-jcc/jjab169
3. Pellino, G., et. al. Delayed diagnosis is influenced by the clinical pattern of Crohn's disease and affects treatment outcomes and quality of life in the long term: a cross-sectional study of 361 patients in Southern Italy. European journal of gastroenterology & hepatology, 27(2), 175–181. https://doi.org/10.1097/MEG.0000000000000244
4. Lee, D., et al. Diagnostic delay in inflammatory bowel disease increases the risk of intestinal surgery. World journal of gastroenterology, 23(35), 6474–6481. https://doi.org/10.3748/wjg.v23.i35.6474
5. Cohen-Mekelburg, S., Berry, S., Stidham, R. W., Zhu, J., & Waljee, A. K. (2021). Clinical applications of artificial intelligence and machine learning-based methods in inflammatory bowel disease. Journal of Gastroenterology and Hepatology, 36(2), 279–285. https://doi.org/10.1111/jgh.15405
6. Peyrin-Biroulet, L. et. al. Defining Disease Severity in Inflammatory Bowel Diseases: Current and Future Directions [Review of Defining Disease Severity in Inflammatory Bowel Diseases: Current and Future Directions]. Perspectives in Clinical Gastroenterology and Hepatology, 14(3), 348–354. https://doi.org/10.1016/j.cgh.2015.06.001
7. Schwab, E., et al, Automatic estimation of ulcerative colitis severity from endoscopy videos using ordinal multi-instance learning. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization. 10. 1-9. 10.1080/21681163.2021.1997644.