Frontiers in oncology · 2026
Artificial intelligence-assisted rapid on-site evaluation in liver biopsy: a diagnostic accuracy study.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
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.
Du Cheng-Cheng, Chen Yu-Xian, Li Chun-Hai, Meng Hong-, Kong Fan-Lei
Аннотация
BACKGROUND: Liver biopsy is the gold standard for diagnosing liver lesions but is hampered by the time-consuming nature of conventional pathology. Artificial Intelligence-assisted Rapid On-Site Evaluation (AI-ROSE) offers a promising solution for real-time assessment of biopsy samples. This study aims to evaluate the diagnostic performance of AI-ROSE in liver biopsies using histopathology as the reference standard.
METHODS: Fifty-eight patients with liver lesions undergoing CT-guided percutaneous biopsy were prospectively enrolled. Each biopsy sample underwent triple assessment: AI-ROSE analysis, exfoliative cytology, and histopathological examination. Diagnostic sensitivity, specificity, accuracy, and predictive values were calculated. Serial (both tests positive) and parallel (either test positive) combinations of AI-ROSE and cytology were also analyzed. The agreement between methods was assessed using the Kappa statistic.
RESULTS: AI-ROSE showed a sensitivity of 92.31% (95% CI: 81.8% - 97.1%) and an accuracy of 87.93% (95% CI: 77.2% - 94.2%), outperforming exfoliative cytology (sensitivity: 86.54%; accuracy: 84.48%). The parallel application of AI-ROSE and cytology achieved the highest sensitivity of 96.15% (95% CI: 87.02-100.00) and significantly improved agreement with the gold standard (Kappa = 0.498, p = 0.0015).
CONCLUSION: This study shows AI-ROSE not only has better sensitivity and accuracy in single-item diagnosis than exfoliative cytology in liver biopsy, but demonstrates excellent synergistic value in parallel application, which can greatly improve the reliability of intraoperative diagnosis. This indicates AI-ROSE has important application prospects in optimizing the clinical decision-making process and reducing the risk of intraoperative missed diagnosis.
METHODS: Fifty-eight patients with liver lesions undergoing CT-guided percutaneous biopsy were prospectively enrolled. Each biopsy sample underwent triple assessment: AI-ROSE analysis, exfoliative cytology, and histopathological examination. Diagnostic sensitivity, specificity, accuracy, and predictive values were calculated. Serial (both tests positive) and parallel (either test positive) combinations of AI-ROSE and cytology were also analyzed. The agreement between methods was assessed using the Kappa statistic.
RESULTS: AI-ROSE showed a sensitivity of 92.31% (95% CI: 81.8% - 97.1%) and an accuracy of 87.93% (95% CI: 77.2% - 94.2%), outperforming exfoliative cytology (sensitivity: 86.54%; accuracy: 84.48%). The parallel application of AI-ROSE and cytology achieved the highest sensitivity of 96.15% (95% CI: 87.02-100.00) and significantly improved agreement with the gold standard (Kappa = 0.498, p = 0.0015).
CONCLUSION: This study shows AI-ROSE not only has better sensitivity and accuracy in single-item diagnosis than exfoliative cytology in liver biopsy, but demonstrates excellent synergistic value in parallel application, which can greatly improve the reliability of intraoperative diagnosis. This indicates AI-ROSE has important application prospects in optimizing the clinical decision-making process and reducing the risk of intraoperative missed diagnosis.