South Dakota medicine : the journal of the South Dakota State Medical Association · 2025
Integrating AI Transcription Software in Medical Education: A Comparative Study of OSCE Documentation.
СтатьяСравнительное исследованиеНазвание 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.
Lien Molly, Max Carson, Fanciullo Rachael, Mortinsen Roy, Schuknecht Arica, Kotalik Mercedes, Kozmenko Valeriy
Аннотация
BACKGROUND: As AI technology becomes increasingly embedded in clinical workflows, medical education must evolve to prepare students for AI-assisted documentation. Freed AI, a real-time transcription tool, offers potential benefits in both efficiency and documentation quality. This study compares the performance of Freed AI-generated notes to student-generated notes across simulated clinical encounters.
METHODS: First-year medical students participated in Objective Structured Clinical Examinations (OSCEs) covering three conditions: cough, falls, and back pain. Each encounter was transcribed both by students and Freed AI. Notes were scored for completeness, accuracy, and medical relevance. Two-way ANOVA and post-hoc tests were conducted to assess differences across source (AI vs. student) and condition.
RESULTS: Freed AI significantly outperformed student-generated notes overall, with a mean score advantage of approximately 3.78 points. A significant interaction was found between source and condition, with AI demonstrating a robust advantage in back pain scenarios. While AI also scored higher in falls and cough, these differences were not statistically significant after correction for multiple comparisons. Condition alone also had a significant effect, with back pain yielding the lowest overall scores.
CONCLUSION: Freed AI transcription significantly enhances documentation quality, particularly in complex scenarios like back pain. These findings support the integration of AI tools into medical education to augment student performance, though continued attention to clinical reasoning and variance in case complexity remains essential.
METHODS: First-year medical students participated in Objective Structured Clinical Examinations (OSCEs) covering three conditions: cough, falls, and back pain. Each encounter was transcribed both by students and Freed AI. Notes were scored for completeness, accuracy, and medical relevance. Two-way ANOVA and post-hoc tests were conducted to assess differences across source (AI vs. student) and condition.
RESULTS: Freed AI significantly outperformed student-generated notes overall, with a mean score advantage of approximately 3.78 points. A significant interaction was found between source and condition, with AI demonstrating a robust advantage in back pain scenarios. While AI also scored higher in falls and cough, these differences were not statistically significant after correction for multiple comparisons. Condition alone also had a significant effect, with back pain yielding the lowest overall scores.
CONCLUSION: Freed AI transcription significantly enhances documentation quality, particularly in complex scenarios like back pain. These findings support the integration of AI tools into medical education to augment student performance, though continued attention to clinical reasoning and variance in case complexity remains essential.