Current problems in diagnostic radiology · 2025
Manuscript title: Impact of vessel suppression AI on reading efficiency and nodule detection in CT chest.
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.
Chorath Kevin T, Jacobs Evan, Tharp Ken, Watt Charles, Bresnahan Brian W, Medverd Jonathan R, Cross Nathan M, Chalian Hamid
Аннотация
BACKGROUND: Vessel suppression (VS) software enhances pulmonary nodule conspicuity by attenuating vascular and fissural structures, but real-world impact on clinical workflows remains uncertain.
METHODS: We retrospectively analyzed 8,835 chest CT examinations (January 2023-February 2024) interpreted at a single academic institution, comparing pre- and post-VS implementation. Reads by residents and faculty were assessed separately. Outcomes included interpretation time and pulmonary nodule detection, extracted using natural language processing with manual validation.
RESULTS: Among residents, mean read time decreased from 19.1 to 12.2 minutes (p<0.001), with improved detection of punctate nodules (13.0% vs 24.2%, p<0.001). Faculty showed reduced mean read time (6.6 vs 5.2 minutes, p<0.001) and modest improvement in punctate nodule detection (26.0% vs 29.5%, p=0.042).
CONCLUSION: VS improved reading efficiency for both trainees and faculty and substantially increased small-nodule detection among residents and attending radiologists. While the clinical significance of detecting sub-2 mm nodules remains uncertain, efficiency gains support the value of VS in high-volume and training environments. Further study is warranted to define downstream clinical impact and guide management of incidentally detected small nodules.
METHODS: We retrospectively analyzed 8,835 chest CT examinations (January 2023-February 2024) interpreted at a single academic institution, comparing pre- and post-VS implementation. Reads by residents and faculty were assessed separately. Outcomes included interpretation time and pulmonary nodule detection, extracted using natural language processing with manual validation.
RESULTS: Among residents, mean read time decreased from 19.1 to 12.2 minutes (p<0.001), with improved detection of punctate nodules (13.0% vs 24.2%, p<0.001). Faculty showed reduced mean read time (6.6 vs 5.2 minutes, p<0.001) and modest improvement in punctate nodule detection (26.0% vs 29.5%, p=0.042).
CONCLUSION: VS improved reading efficiency for both trainees and faculty and substantially increased small-nodule detection among residents and attending radiologists. While the clinical significance of detecting sub-2 mm nodules remains uncertain, efficiency gains support the value of VS in high-volume and training environments. Further study is warranted to define downstream clinical impact and guide management of incidentally detected small nodules.