Journal of the American College of Radiology : JACR · 2026
A Document-Grounded Artificial Intelligence Chatbot as a Bridge to Mentorship in Radiology Workforce Onboarding: A Qualitative 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.
Tan Celine Ying Yi, Chong Letty Ai Jia, Chin Crystal Jing, Dong Chaoyan
Аннотация
PURPOSE: Mentorship is central to radiology onboarding, yet increasing trainee-to-preceptor ratios and variable educator availability can result in inconsistent learning experiences for newly recruited radiologic technologists. This study examined how a document-grounded artificial intelligence (AI) chatbot supported clinical onboarding and complemented workplace mentorship.
METHODS: Twenty-one newly appointed radiologic technologists who used an institutionally developed document-grounded AI chatbot (XPA) during a 12-week onboarding program participated in three semistructured online focus group discussions. XPA was built on a government AI platform and configured to generate responses exclusively from uploaded institution-specific reference documents, including clinical protocols and workflow guides. Data were analyzed using hybrid deductive-inductive thematic analysis.
RESULTS: Three themes emerged. XPA improved access to institution-specific knowledge by enabling rapid, standardized retrieval of protocol and workflow information previously distributed across intranet resources. Trust in XPA was task dependent: Participants relied on it for foundational learning and routine protocol clarification but deferred to senior radiologic technologists for contextual and adaptive clinical decisions. XPA also lowered psychological barriers to help-seeking by enabling private, non-judgemental questioning, functioning as a rehearsal space that prepared participants for more focused and confident engagement with senior mentors.
CONCLUSIONS: Document-grounded AI can serve as a preparatory scaffold during radiology onboarding, improving knowledge accessibility and supporting help-seeking behaviors without replacing clinical mentorship. A human-centered implementation framework is proposed, positioning AI as a complement to-rather than a substitute for-senior mentorship and clinical judgement.
METHODS: Twenty-one newly appointed radiologic technologists who used an institutionally developed document-grounded AI chatbot (XPA) during a 12-week onboarding program participated in three semistructured online focus group discussions. XPA was built on a government AI platform and configured to generate responses exclusively from uploaded institution-specific reference documents, including clinical protocols and workflow guides. Data were analyzed using hybrid deductive-inductive thematic analysis.
RESULTS: Three themes emerged. XPA improved access to institution-specific knowledge by enabling rapid, standardized retrieval of protocol and workflow information previously distributed across intranet resources. Trust in XPA was task dependent: Participants relied on it for foundational learning and routine protocol clarification but deferred to senior radiologic technologists for contextual and adaptive clinical decisions. XPA also lowered psychological barriers to help-seeking by enabling private, non-judgemental questioning, functioning as a rehearsal space that prepared participants for more focused and confident engagement with senior mentors.
CONCLUSIONS: Document-grounded AI can serve as a preparatory scaffold during radiology onboarding, improving knowledge accessibility and supporting help-seeking behaviors without replacing clinical mentorship. A human-centered implementation framework is proposed, positioning AI as a complement to-rather than a substitute for-senior mentorship and clinical judgement.