Deep Learning-Based Confocal Laser Microendoscopy Feature Atlas Construction and Its Application in Intelligent Diagnosis of Irritable Bowel Syndrome

Overview

In this study, the healthy control group and IBS patients are taken as the research subjects, and CLE is applied to image and analyze the enrolled persons respectively, to derive and compare the characteristic microstructural images of the healthy control group with those of the IBS patients, to establish a diagnostic model of IBS by using deep learning and to evaluate the diagnostic efficacy of the model for IBS.Participants will undergo colonoscopy and confocal laser microendoscopy, and IBS patients will undergo targeted biopsy in CLE to observe suspicious lesions and lesion margins, and specimens will undergo HE staining and immunohistochemistry testing. Translated with DeepL.com (free version)

Detailed description

The study is prospective, observational, and diagnostic.

Research methodology Equipment Consumables The equipment consumables used include a colonoscopy operating system, pCLE system, fluorescein sodium injection, and single-use endoscopic biopsy forceps.

Phase 1 Endoscopy All enrolled person receive fluorescein sodium allergy test before pCLE examination, and fluorescein sodium is injected intravenously after observing no allergic reactions such as dizziness, rash, itching, shock, etc. for 15 min. 30 s later, the pCLE probe is placed from the biopsy clamp channel of enteroscopy, and six different intestinal segments of the cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum are examined in order and the images are collected, and each intestinal segment of the pCLE of all person is recorded continuously with 3-minute video recordings (operation is kept within 30 min after the procedure). and more than 4000 images are recorded for each person. During the operation, the probe is gently placed on the mucosal surface and kept as perpendicular to the bowel wall as possible, and the movements are gentle, slow, and soft to avoid mucosal injury. The endoscopist is performing targeted biopsies of all suspicious lesions and lesion margins observed in pCLE.

Pathological analysis After the specimens are fixed in formalin solution and embedded in paraffin, 2 sections with a thickness of 4 μm are cut. The first section is stained with HE to evaluate the mild inflammation of the intestine; the second section is stained with immunohistochemistry to evaluate the mast cell markers and tight junction proteins in the intestinal wall.

Screening of characteristic pCLE images All videos are measured and analyzed by the same experienced endoscopist, after screening all images with good stability, no artifacts, and clear display of microstructures. pCLE images are characterized by a ruler of 10 μm in length in the lower left corner, which allows real-time evaluation of a number of parameters in real time during the operation. Image J image analysis software is used to measure capillary diameter (CD), cell spacing (CS), gland spacing (GS), gland area (GA), and whether sodium fluorescein leak into the crypt lumen in the pCLE field of view. and other indicators. The images are categorized according to the healthy control group and the IBS patient group, and the characteristic images of each group are formed.

Phase 2 Diagnostic model construction of IBS patient images Deep learning is performed based on two sets of characteristic images to construct a diagnostic model for IBS.

Model evaluation Patients with healthy, IBS, early IBD, or celiac disease (total sample size of 30 cases for total diagnostic model efficacy validation in phases 2 and 3) are examined for pCLE in the same way as in phase 1, and high-quality images are screened for discriminative diagnosis of this model and evaluated for accuracy and inter-evaluator agreement of their diagnostic findings.

Phase 3 Construction of a multimodal diagnostic model for images of IBS patients with joint text On the basis of phase 2, the text data of various indicators corresponding to the images are added for joint deep learning to construct a multimodal diagnostic model for IBS patients.

Model evaluation Patients with healthy, IBS, early IBD, or celiac disease (total sample size of 30 cases for validation of diagnostic model efficacy in phases 2 and 3 combined) are examined for pCLE in the same way as in phase 1, with high-quality images and various types of indicators screened for discriminatory diagnosis of this model and evaluated for the accuracy of the diagnostic results and inter-appraisal agreement.

Primary outcome measures

• Evaluation of the accuracy and inter-evaluator agreement of the constructed IBS diagnostic model with the diagnostic results of the Rome IV criteria. [Time frame: Approximately 15 months from completion of image collection, analysis, deep learning and construction of IBS diagnostic model to validation of diagnostic model performance.]

Eligibility criteria

Inclusion criteria

• ①Healthy control group and test group of IBS with clear clinical diagnosis (Rome IV diagnostic criteria). Healthy controls are free of gastrointestinal symptoms, test results (blood count, CRP, ESR, etc.) and colonoscopy abnormalities.Patients with IBS have recurrent abdominal pain (≥1 day per week on average in the past 3 months), with onset ≥6 months before diagnosis, and the abdominal pain is accompanied by at least two of the following three symptoms: pain related to defecation, change in stool frequency, and change in the shape (appearance) of the feces, and other organic or metabolic disorders are excluded.② Age: between 18 and 80 years old.③ Probe-based confocal Laser Endomicroscopy (pCLE) is required, and patients have good compliance. ④ Normal coagulation function, and biopsies can be obtained for pathological examination.⑤ Voluntary signing of informed consent.

Exclusion criteria

• Fluorescein sodium allergy. ②Contraindications to colonoscopy (pregnant or breastfeeding patients, uncooperative psychiatric patients with severe mental disorders, colonic obstruction) or inadequate bowel preparation (Boston Bowel Preparedness Scale score of <2 for any colonic segment).③ Comorbidities with major organ dysfunction (with severe cardiopulmonary disease) or impaired renal function (blood creatinine >450umol/L). ④ Other conditions that, in the judgment of the investigator, make it unsuitable for participation in the study (inflammatory bowel disease, infected IBS, having taken nonsteroidal anti inflammatory drugs, corticosteroids or antibiotics within the last year, etc.).

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: Yes

Study design

Observational model

Other

Study locations

Center list to be confirmed — check the primary protocol.

Publications

• Fritscher-Ravens A, Pflaum T, Mosinger M, Ruchay Z, Rocken C, Milla PJ, Das M, Bottner M, Wedel T, Schuppan D. Many Patients With Irritable Bowel Syndrome Have Atypical Food Allergies Not Associated With Immunoglobulin E. Gastroenterology. 2019 Jul;157(1):109-118.e5. doi: 10.1053/j.gastro.2019.03.046. Epub 2019 May 15. PMID 31100380
• Fritscher-Ravens A, Schuppan D, Ellrichmann M, Schoch S, Rocken C, Brasch J, Bethge J, Bottner M, Klose J, Milla PJ. Confocal endomicroscopy shows food-associated changes in the intestinal mucosa of patients with irritable bowel syndrome. Gastroenterology. 2014 Nov;147(5):1012-20.e4. doi: 10.1053/j.gastro.2014.07.046. Epub 2014 Jul 30. PMID 25083606
• Bertiaux-Vandaele N, Youmba SB, Belmonte L, Lecleire S, Antonietti M, Gourcerol G, Leroi AM, Dechelotte P, Menard JF, Ducrotte P, Coeffier M. The expression and the cellular distribution of the tight junction proteins are altered in irritable bowel syndrome patients with differences according to the disease subtype. Am J Gastroenterol. 2011 Dec;106(12):2165-73. doi: 10.1038/ajg.2011.257. Epub 2011 PMID 22008894
• Liu JJ, Rudzinski JK, Mah SJ, Thiesen AL, Bao H, Wine E, Ogg SC, Boulanger P, Fedorak RN, Madsen KL. Epithelial gaps in a rodent model of inflammatory bowel disease: a quantitative validation study. Clin Transl Gastroenterol. 2011 Jun 9;2(6):e3. doi: 10.1038/ctg.2011.2. PMID 23237881
• Robles-Medranda C, Oleas R, Valero M, Puga-Tejada M, Soria-Alcivar M, Ospina J, Alvarado-Escobar H, Munoz-Jurado G, Baquerizo-Burgos J, Pitanga-Lukashok H. Confocal laser endomicroscopy detects colonic inflammation in patients with irritable bowel syndrome: a prospective study. Endosc Int Open. 2020 Apr;8(4):E550-E557. doi: 10.1055/a-1119-6327. Epub 2020 Mar 23. PMID 32258379
• Kiesslich R, Goetz M, Angus EM, Hu Q, Guan Y, Potten C, Allen T, Neurath MF, Shroyer NF, Montrose MH, Watson AJ. Identification of epithelial gaps in human small and large intestine by confocal endomicroscopy. Gastroenterology. 2007 Dec;133(6):1769-78. doi: 10.1053/j.gastro.2007.09.011. Epub 2007 Sep 16. PMID 18054549
• Liu JJ, Wong K, Thiesen AL, Mah SJ, Dieleman LA, Claggett B, Saltzman JR, Fedorak RN. Increased epithelial gaps in the small intestines of patients with inflammatory bowel disease: density matters. Gastrointest Endosc. 2011 Jun;73(6):1174-80. doi: 10.1016/j.gie.2011.01.018. Epub 2011 Mar 11. PMID 21396639
• ASGE Technology Committee. Confocal laser endomicroscopy. Gastrointest Endosc. 2014 Dec;80(6):928-38. doi: 10.1016/j.gie.2014.06.021. No abstract available. PMID 25442092

Identifiers

NCT: NCT07051226 · Ratification NO:2025(203)

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗