O'Rourke Sean, Xu Sophia, Carrero Stephanie, Drebin Harrison M, Felman Ariel, Ko Andrew, Misseldine Adam, Mouchtaris Sofia G, Musialowicz Brett, Wong Tony T, Zech John R
Department of Radiology, Columbia University Irving Medical Center, 622 W 168 St, New York, NY, 10032, USA.
Department of Medical Education, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, USA.
Skeletal Radiol. 2025 Apr 14. doi: 10.1007/s00256-025-04927-0.
Prior work has demonstrated that AI access can help residents more accurately detect pediatric fractures. We wished to evaluate the effectiveness of an unsupervised AI-based training module as a pediatric fracture detection educational tool.
Two hundred forty radiographic examinations from throughout the pediatric upper extremity were split into two groups of 120 examinations. A previously developed open-source deep learning fracture detection algorithm ( www.childfx.com ) was used to annotate radiographs. Four medical students and four PGY-2 radiology residents first evaluated 120 examinations for fracture without AI assistance and subsequently reviewed AI annotations on these cases via a training module. They then interpreted 120 different examinations without AI assistance. Pre- and post-intervention fracture detection accuracy was evaluated using a chi-squared test.
Overall resident fracture detection accuracy significantly improved from 71.3% pre-intervention to 77.5% post-intervention (p = 0.032). Medical student fracture detection accuracy was not significantly changed from 56.3% pre-intervention to 57.3% post-intervention (p = 0.794). Eighty-eight percent of responding participants (7/8) would recommend this model of learning.
We found that a tailored AI-based training module increased resident accuracy for detecting pediatric fractures by 6.2%. Medical student accuracy was not improved, likely due to their limited background familiarity with the task. AI offers a scalable method for automatically generating annotated teaching cases covering varied pathology, allowing residents to efficiently learn from simulated experience.
先前的研究表明，使用人工智能辅助可以帮助住院医师更准确地检测小儿骨折。我们希望评估一个基于人工智能的无监督训练模块作为小儿骨折检测教育工具的有效性。
从整个小儿上肢的240份X光检查中分为两组，每组120份检查。使用先前开发的开源深度学习骨折检测算法（www.childfx.com）对X光片进行标注。四名医学生和四名放射科二年级住院医师首先在没有人工智能辅助的情况下评估120份检查是否存在骨折，随后通过一个训练模块查看这些病例的人工智能标注。然后他们在没有人工智能辅助的情况下解读另外120份不同的检查。使用卡方检验评估干预前后骨折检测的准确性。
住院医师总体骨折检测准确率从干预前的71.3%显著提高到干预后的77.5%（p = 0.032）。医学生骨折检测准确率从干预前的56.3%到干预后的57.3%没有显著变化（p = 0.794）。88%的受访者（7/8）会推荐这种学习模式。
我们发现，一个量身定制的基于人工智能的训练模块将住院医师检测小儿骨折的准确率提高了6.2%。医学生的准确率没有提高，可能是因为他们对这项任务的背景熟悉程度有限。人工智能提供了一种可扩展的方法，用于自动生成涵盖各种病理情况的标注教学病例，使住院医师能够从模拟经验中高效学习。
