Real-time AI-assisted Endocyroscopy for the Diagnosis of Colorectal Lesions

Real-time AI-assisted Endocytoscopy for the Diagnosis of Colorectal Lesions: a Multi-center, Prospective Clinical Study

Colorectal cancer (CRC) is the third most common malignancy and the second leading cause of cancer-related death worldwide. Colonoscopy is considered the preferred method of screening for colorectal cancer, and early and resection detection of colorectal neoplastic lesions can significantly reduce colorectal cancer morbidity and mortality. In order to improve the diagnostic accuracy of endoscopy for colorectal lesions, many endoscopic techniques, such as image-enhanced endoscopy, including narrow band imaging (narrow-band imaging, NBI), magnifying endoscopy, pigment endoscopy, confocal laser endoscopy, and endocytoscopy(EC), are applied clinically. However, with the increasing number of endoscopic resection, the costs associated with the pathological diagnosis of endoscopic resection and resection specimens increase year by year. In clinical practice, some non-neoplastic colorectal lesions may not require resection, so it is important to identify the nature of the lesion during colonoscopy. Leveraging deep neural networks, AI systems support both computer-aided detection (CADe) and computer-aided classification (CADx). CADe specifically focuses on identifying polyps in colonoscopy, with the goal of reducing adenoma miss rates. Hovever, CADx can predict the pathology of the lesion based on the surface condition of the lesion. Endocytoscopy is a kind of ultra-high magnification endoscopy. But it is not something that can be easily mastered by endoscopic doctors. The investigators have previously developed an artificial intelligence system that can assist in endocytoscopy. The investigators plan to conduct a prospective, multicenter clinical trial to verify the accuracy of this CADx in predicting the histological characteristics of colorectal lesions during real-time endocytoscopy.

Study Overview

• Status: Completed
• Conditions: Colorectal Neoplasms; Colorectal Polyp
• Intervention / Treatment: Diagnostic test: Computer-aided diagnosis (CADx) support tool

Detailed Description

Colonoscopy is currently the gold standard of screening for CRC. Endocytoscopy is a kind of ultra-high magnification endoscopy. Combined with chemical staining and narrow band imaging technology, endoscopists can observe the cell nucleus morphology, gland tube morphology and microvascular morphology with the naked eye, so as to avoid pathological examination and realize the purpose of real-time biopsy in the body. However, the judgment of endocytoscopic images needs a lot of experience to improve the judgment accuracy. Moreover, endoscopists have certain subjective judgments and errors in the process of judging the results. Therefore, artificial intelligence (AI) is proposed for computer-assisted diagnosis in clinic to solve this problem. At present, the available artificial intelligence systems for assisting endoscopists in diagnosing using endocytoscopy are still relatively scarce, and they are based on traditional machine learning methods, which still have certain limitations in terms of accuracy. With the continuous development of computer vision technology, deep learning has been widely applied in the development of endoscopic assistance diagnosis systems. Therefore, the investigators developed a CADx system trained using deep learning to assist endoscopists in making diagnoses when using endocytoscopy. This CADx system can predict the captured endocytoscopy images in real time and display the prediction results, which can assist endoscopists in providing diagnostic references.

However, currently this CADx technology has not yet undergone prospective clinical validation in the clinical setting. The investigators plan to conduct a prospective clinical trial to validate the accuracy of CADx for prediction of colorectal lesions histology in real-time endocytoscopy. This study will prospectively collect the lesions that meet the inclusion and exclusion criteria. After the endoscopists make the diagnosis through endoscopic images and CADx and then undergo endoscopic resection or surgical resection followed by pathological diagnosis, they will compare the artificial intelligence diagnosis results with the gold standard pathological results, and summarize the diagnostic accuracy of this artificial intelligence-assisted diagnostic system for the colorectal lesions.

The investigators plan to conduct a prospective, multi-centre clinical trial to validate the accuracy of CADx support for prediction of colorectal lesions histology in real-time endocytoscopy.

• Study Type: Observational
• Enrollment (Actual): 570

Contacts and Locations

Study Locations

• China
  • Jilin
    • Changchun, Jilin, China, 130021
      • The First Hospital of Jilin University
    • Meihekou, Jilin, China, 135000
      • Meihekou Central Hospital
  • Shandong
    • Jinan, Shandong, China, 250000
      • Shandong Second Provincial General Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

Patients with one or more colorectal lesions detected during endocytoscopy will be included in the study. The rest of the inclusion and exclusion criteria are as described.

Description

Inclusion Criteria:

• Patients who have at least one colorectal lesion detected during endocytoscopy
• Consent obtained for the study

Exclusion Criteria:

• lesions lacking high-quality images;
• Inflammatory bowel disease, familial adenomatous polyposis and other special diseases;
• Submucosal tumors;
• Pathological diagnosis of inflammatory polyps, Peutz-Jeghers polyps, juvenile polyps, lymphoma and other special pathological types.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Patients with one or more colorectao lesion detected; During colonoscopy, the endoscopist inspect for the presence of colorectal lesions as per routine clinical practice with the CADx turned off. When a colorectal lesion is encountered, the endoscopist will make a prediction on the histology based on the endoscopic diagnosis. Following this, the CADx will be triggered and display the endoscopic image captured by the endoscopist. and the endoscopist will take note of the CADx prediction for the same image. In addition, other lesion features such as the size and location will be recorded, which is similar to what is performed in routine clinical practice. The lesion will be endoscopic resected or surgery and sent for pathological examination, which will form the "gold standard" for the diagnosis of polyp histology.

Diagnostic test: Computer-aided diagnosis (CADx) support tool; The CADx support tool will display the prediction results when the endoscopists press the keys on the fixed keyboard. This is performed after the endoscopists first makes an optical prediction of colorectal lesion histology using endocytoscopy as described. The CADx support tool will make a prediction of colorectal lesion histology.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
To evaluate the diagnostic performance of the CAD-stained in diagnosing neoplastic lesions in a clinical setting.	The diagnostic performance will be calculated for comparison with final histology as the gold standard for diagnosis	11 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
To evaluate the diagnostic performance of the CAD-NBI in diagnosing neoplastic lesions in a clinical setting.	The diagnostic performance will be calculated for comparison with final histology as the gold standard for diagnosis	11 months
To compare the diagnostic performance of CAD-NBI and CAD-stained in colorectal neoplastic lesions different colorectal lesions.	The diagnostic performance of the CAD-NBI and CAD-stained will be calculated for comparison with final histology as the gold standard for diagnosis	11 months
To evaluate the diagnostic performance of CAD-NBI and CAD-stained in the diagnosis of neoplastic DRSPs with high confidence	The diagnostic performance of CAD-NBI and CAD-stained will be calculated for comparison with final histology as the gold standard for diagnosis	11 months
To evaluate the agreement of post-polypectomy surveillance intervals based on CAD-NBI and CAD-stained predictions with histopathological diagnosis	The agreement of post-polypectomy surveillance intervals based on CAD-NBI and CAD-stained predictions with The pathological diagnosis was made according to the guidelines	11 months

Collaborators and Investigators

• Sponsor: The First Hospital of Jilin University
• Collaborators: Meihekou Central Hospital; Shandong Second Provincial General Hospital
• Investigators: Principal Investigator: Hong Xu, Docror, The First Hospital of Jilin University

Study record dates

Study Major Dates

• Study Start (Actual): February 5, 2025
• Primary Completion (Actual): December 29, 2025
• Study Completion (Actual): December 29, 2025

Study Registration Dates

• First Submitted: January 20, 2025
• First Submitted That Met QC Criteria: January 20, 2025
• First Posted (Actual): January 24, 2025

Study Record Updates

• Last Update Posted (Actual): April 13, 2026
• Last Update Submitted That Met QC Criteria: April 8, 2026
• Last Verified: April 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Neoplasms by Site; Neoplasms; Intestinal Diseases; Gastrointestinal Neoplasms; Digestive System Neoplasms; Digestive System Diseases; Gastrointestinal Diseases; Intestinal Neoplasms; Rectal Diseases; Colonic Diseases; Colorectal Neoplasms; Diagnosis; Diagnosis, Computer-Assisted
• Other Study ID Numbers: 25K013-001

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No