Digestive endoscopy : official journal of the Japan Gastroenterological Endoscopy Society · 2026
Diagnostic Performance and Reproducibility of Chromoendoscopic Classification (FOCUS Classification: Flat-Type Dysplasia Optical Assessment by Chromoendoscopy in Ulcerative ColitiS) for Flat-Type Dysplasia in Ulcerative Colitis-Associated Neoplasia.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Takabayashi Kaoru, Murata Shoma, Sasaki Motoki, Miyazaki Kurato, Masunaga Teppei, Kirita Kumiko, Mizutani Mari, Yoshimatsu Yusuke, Takatori Yusaku, Akimoto Teppei, Kiyohara Hiroki, Sugimoto Shinya, Kawasaki Shintaro, Mikami Yohei, Matsuura Noriko, Nakayama Atsushi, Sujino Tomohisa, Ito Yasuhiro, Yamanoi Kazuhiro, Sekine Shigeki, Iwao Yasushi, Yahagi Naohisa, Kanai Takanori, Kato Motohiko
Аннотация
OBJECTIVES: Ulcerative colitis-associated neoplasia (UCAN) often presents as flat-type dysplasia that is difficult to characterize endoscopically. We developed a FOCUS (Flat-type dysplasia Optical assessment by Chromoendoscopy in Ulcerative colitiS) classification and evaluated its diagnostic performance and reproducibility.
METHODS: This single-center retrospective study included 33 UCAN lesions from 26 patients. Prospectively recorded FOCUS classifications at 230 biopsy sites were compared with histopathology. Chromoendoscopic patterns were categorized as Type I ("suggestive of non-dysplasia,"), Type II ("suggestive of dysplasia with low confidence"), or Type III ("suggestive of dysplasia with high confidence"). Diagnostic performance was assessed using two prespecified thresholds: Types II/III versus Type I and Type III versus Types I/II. Eight endoscopists independently classified all images; interobserver and intraobserver agreement were assessed using Fleiss' and Cohen's κ, respectively.
RESULTS: The dysplasia-bearing rate increased stepwise from Type I to Type III (Type I 15.5%, Type II 58.3%, Type III 81.8%). Using the primary threshold, sensitivity was 82.0%, and specificity 75.4% (positive likelihood ratio 3.33; negative likelihood ratio 0.24). Interobserver agreement for the three-category classification was moderate to substantial (Fleiss' κ 0.629; 95% CI 0.545-0.712), and median intraobserver agreement was substantial (Cohen's κ 0.738).
CONCLUSIONS: The FOCUS classification provides clinically meaningful risk stratification for flat-type dysplasia in a high-risk UCAN cohort, with acceptable interobserver and intraobserver reproducibility.
METHODS: This single-center retrospective study included 33 UCAN lesions from 26 patients. Prospectively recorded FOCUS classifications at 230 biopsy sites were compared with histopathology. Chromoendoscopic patterns were categorized as Type I ("suggestive of non-dysplasia,"), Type II ("suggestive of dysplasia with low confidence"), or Type III ("suggestive of dysplasia with high confidence"). Diagnostic performance was assessed using two prespecified thresholds: Types II/III versus Type I and Type III versus Types I/II. Eight endoscopists independently classified all images; interobserver and intraobserver agreement were assessed using Fleiss' and Cohen's κ, respectively.
RESULTS: The dysplasia-bearing rate increased stepwise from Type I to Type III (Type I 15.5%, Type II 58.3%, Type III 81.8%). Using the primary threshold, sensitivity was 82.0%, and specificity 75.4% (positive likelihood ratio 3.33; negative likelihood ratio 0.24). Interobserver agreement for the three-category classification was moderate to substantial (Fleiss' κ 0.629; 95% CI 0.545-0.712), and median intraobserver agreement was substantial (Cohen's κ 0.738).
CONCLUSIONS: The FOCUS classification provides clinically meaningful risk stratification for flat-type dysplasia in a high-risk UCAN cohort, with acceptable interobserver and intraobserver reproducibility.