Clinical Study on the Diagnosis of Colorectal Lesions by Real-time Artificial Intelligence Assisted Endocytoscopy Combined With Narrow Band Imaging

Обзор

Colorectal cancer (colorectal cancer, CRC) is the third most common malignant tumor globally and the second leading cause of cancer-related deaths. Colonoscopy is considered the preferred method for screening colorectal cancer; early detection and removal of colorectal neoplasms can significantly reduce the incidence and mortality of colorectal cancer. To improve the diagnostic accuracy of endoscopy in colorectal lesions, many endoscopic techniques have been applied clinically, such as image-enhanced endoscopy, including narrow band imaging (narrow-band imaging, NBI), magnifying endoscopy, chromoendoscopy, confocal laser endoscopy, and endocytoscopy (EC). However, with the increasing number of endoscopic resections, the costs associated with the pathological diagnosis of resected specimens have risen year by year. In clinical practice, some non-neoplastic colorectal lesions may not require resection, so it is important to differentiate the nature of lesions during colonoscopy. Endocytoscopy is an ultra-high magnification endoscope that, when combined with chemical staining and narrowband imaging techniques, allows the endoscopist to observe the nuclear morphology of colorectal lesions, the shape of glands, and the morphology of microvessels through the naked eye. This approach avoids the need for pathological examination, achieving the goal of real-time biopsy in vivo. However, the accuracy of endocytoscopic image interpretation requires extensive experience to improve judgment, and there is a certain degree of subjectivity and error in the endoscopist's assessment process. Therefore, to address this issue, clinical applications have proposed using artificial intelligence (AI) for computer-aided diagnosis. The investigator's center has previously developed an AI-assisted diagnostic system based on endocytoscopy with NBI to assist in determining the nature of colorectal lesions. However, forward-looking clinical studies are still lacking to verify the effectiveness of this AI-assisted system. Thus, the investigator aim to conduct such clinical research to validate the clinical efficacy of this AI.

Вмешательства

• Диагностический тест artificial intelligence
  Artificial intelligence assisted diagnostic system was used to diagnose colorectal lesions

Первичные конечные точки

• The accuracy of AI in diagnosing tumor lesions (including sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV)) and high confidence rate were evaluated [Срок оценки: 2025-12-31]

Вторичные конечные точки (6)

• The accuracy and high confidence rate of AI in diagnosing non-neoplastic lesions, adenomas and invasive cancers were evaluated [Срок оценки: 2025-12-31]
• The accuracy and high confidence rate of AI diagnosis of rectosigmoid adenomas ≤5 mm were evaluated [Срок оценки: 2025-12-31]
• The influence of lesion location, size and shape on artificial intelligence diagnosis of lesion nature was evaluated. [Срок оценки: 2025-12-31]
• The accuracy and high confidence rate of artificial intelligence, endoscopists and endoscopists combined with artificial intelligence in diagnosing lesion nature were compared [Срок оценки: 2025-12-31]
• Compare the time it takes for an endoscopist and an AI to make a diagnosis [Срок оценки: 2025-12-31]
• To compare whether the diagnostic efficacy of the intracellular endoscopic AI-assisted diagnosis system for diagnosing colorectal neoplastic lesions is not inferior to that of the EndoBRAIN-NBI model. [Срок оценки: 2025-12-31]

Критерии участия

Критерии включения

• colorectal lesions

Критерии исключения

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

Критерии приведены из реестра в оригинале (на английском). Окончательную оценку соответствия проводит исследовательский центр.

Дизайн исследования

Модель наблюдения

Только случаи

Центры проведения

Китай · 1 центр

• First Hospital of Jilin University — Changchun

Идентификаторы

NCT: NCT06982885 · 25K190-001

Первоисточники (государственные реестры)

• ClinicalTrials.gov (NIH, США)

Открыть это исследование на ClinicalTrials.gov ↗