• Clinical Endoscopy
• /
• 제55권1호
• /
• pp.113-121
• /
• 2022

Background/Aims: We have been developing artificial intelligence based polyp histology prediction (AIPHP) method to classify Narrow Band Imaging (NBI) magnifying colonoscopy images to predict the hyperplastic or neoplastic histology of polyps. Our aim was to analyze the accuracy of AIPHP and narrow-band imaging international colorectal endoscopic (NICE) classification based histology predictions and also to compare the results of the two methods. Methods: We studied 373 colorectal polyp samples taken by polypectomy from 279 patients. The documented NBI still images were analyzed by the AIPHP method and by the NICE classification parallel. The AIPHP software was created by machine learning method. The software measures five geometrical and color features on the endoscopic image. Results: The accuracy of AIPHP was 86.6% (323/373) in total of polyps. We compared the AIPHP accuracy results for diminutive and non-diminutive polyps (82.1% vs. 92.2%; p=0.0032). The accuracy of the hyperplastic histology prediction was significantly better by NICE compared to AIPHP method both in the diminutive polyps (n=207) (95.2% vs. 82.1%) (p<0.001) and also in all evaluated polyps (n=373) (97.1% vs. 86.6%) (p<0.001) Conclusions: Our artificial intelligence based polyp histology prediction software could predict histology with high accuracy only in the large size polyp subgroup.

• Journal of Digestive Cancer Research
• /
• 제12권2호
• /
• pp.90-101
• /
• 2024

The incidence of colon cancer in South Korea has recently been the highest among gastrointestinal cancers. Early diagnosis is critical, and image-enhanced endoscopy (IEE) is a key diagnostic method. Colon tumors primarily include serrated polyps, adenomatous polyps, and colon cancer. Early endoscopic techniques relied on simple visual inspection for diagnosis, with tumor size and shape being the primary considerations. Low-resolution images made these methods ineffective for detecting small or early-stage lesions. IEE now enables detailed examination using high-resolution images and various color and structure analyses. Techniques like narrow band imaging (NBI) allow precise observation of vascular patterns and surface structures. Hyperplastic polyps often appear similar in color to the surrounding mucosa, with no visible vascular pattern. Sessile serrated lesions have a cloudy surface with distinct boundaries and irregular patterns, often with black spots in the crypts. Adenomatous polyps are darker brown, with a visible white epithelial network and various pit patterns. Magnified images help differentiate between low- and high-grade dysplasia, with low-grade showing regular patterns and high-grade showing increased irregularities. The NBI International Colorectal Endoscopic classification identifies malignant colon tumors as brown or dark brown with disorganized vascular patterns. The Japan NBI Expert Team classification includes loose vascular areas and disrupted thick vessels. The Workgroup serrAted polypS and Polyposis classification aids in differentiating between hyperplastic polyps and sessile serrated lesions/adenomas when deciding whether to resect polyps larger than 5 mm. Suspected high-grade dysplasia warrants endoscopic submucosal dissection and follow-up. Future advancements in IEE are expected to further enhance early detection and diagnostic accuracy.

• Clinical Endoscopy
• /
• 제55권5호
• /
• pp.645-654
• /
• 2022

Background/Aims: Narrow Band Imaging (NBI) International Colorectal Endoscopic (NICE) and Workgroup Serrated Polyps and Polyposis (WASP) classifications were developed for optical diagnosis of neoplastic and sessile serrated polyps, respectively. Near-focus NBI with NICE combined with WASP criteria for optical diagnosis of colonic polyps has not yet been evaluated. We aimed to compare the accuracy of near-focus NBI (group A) with normal-focus NBI (group B) in real-time optical diagnosis of colorectal polyps using combined NICE and WASP criteria. Methods: Among 362 patients, 118 with 227 polyps were recruited. Groups A and B included 62 patients with 130 polyps (three lost polyps) and 56 patients with 106 polyps (six lost polyps), respectively. Optical diagnoses were compared with pathological reports. Results: The accuracy of optical diagnosis of neoplastic polyps in groups A and B was not significantly different (76% vs. 71%, p=0.52). WASP criteria provided all false positive diagnoses of sessile polyps as serrated polyps in 31 (16.2%) patients. Conclusions: Near-focus NBI was not superior to normal-focus NBI in optical diagnostics of neoplastic polyps using NICE criteria. In our study, WASP classification yielded all false positives in the diagnosis of sessile serrated adenomas/polyps. Routine real-life optical diagnosis of polyps is still unadvisable.