Department of Obstetrics and Prenatal Medicine, University Hospital Bonn, Venusberg Campus 1, 53127, Bonn, Germany.
Office of Academic Affairs, University Hospital Bonn, Venusberg Campus 1, 53127, Bonn, Germany.
Arch Gynecol Obstet. 2024 Oct;310(4):2055-2064. doi: 10.1007/s00404-024-07561-8. Epub 2024 May 25.
We used two 3D ultrasound volumes of fetal heads at 13 weeks to create live-size 3D-printed phantoms with a view to training or assessment of diagnostic abilities for normal and abnormal nuchal translucency measurements. The phantoms are suitable for use in a water bath, imitating a real-life exam. They were then used to study measurement accuracy and reproducibility in examiners of different skill levels.
Ultrasound scans of a 13 + 0-week fetus were processed using 3D Slicer software, producing a stereolithography file for 3D printing. The model, crafted in Autodesk Fusion360™, adhered to FMF guidelines for NT dimensions (NT 2.3 mm). Additionally, a model with pathologic NT was designed (NT 4.2 mm). Printing was performed via Formlabs Form 3® printer using High Temp Resin V2. The externally identical looking 3D models were embedded in water-filled condoms for ultrasound examination. Eight specialists of varying expertise levels conducted five NT measurements for each model, classifying them in physiological and abnormal models.
Classification of the models in physiological or abnormal NT resulted in a detection rate of 100%. Average measurements for the normal NT model and the increased NT model were 2.27 mm (SD ± 0.38) and 4.165 mm (SD ± 0.51), respectively. The interrater reliability was calculated via the intraclass correlation coefficient (ICC) which yielded a result of 0.883, indicating robust agreement between the raters. Cost-effectiveness analysis demonstrated the economical nature of the 3D printing process.
This study underscores the potential of 3D printed fetal models for enhancing ultrasound training through high inter-rater reliability, consistency across different expert levels, and cost-effectiveness. Limitations, including population variability and direct translation to clinical outcomes, warrant further exploration. The study contributes to ongoing discussions on integrating innovative technologies into medical education, offering a practical and economical method to acquire, refine and revise diagnostic skills in prenatal ultrasound. Future research should explore broader applications and long-term economic implications, paving the way for transformative advancements in medical training and practice.
我们使用了两个 13 周胎儿的三维超声容积，创建了与实物大小相当的 3D 打印模型，旨在培训或评估正常和异常颈项透明层（NT）测量的诊断能力。这些模型适合在水浴中使用，模拟真实的检查。然后，我们使用这些模型来研究不同技能水平的检查者的测量准确性和可重复性。
使用 3D Slicer 软件对 13+0 周胎儿的超声扫描进行处理，生成用于 3D 打印的立体光刻文件。该模型由 Autodesk Fusion360TM 制作，符合 NT 尺寸的 FMF 指南（NT 2.3mm）。此外，还设计了一个有病理 NT 的模型（NT 4.2mm）。通过 Formlabs Form 3®打印机使用高温树脂 V2 进行打印。外部看起来相同的 3D 模型被嵌入充满水的避孕套中进行超声检查。八位不同专业水平的专家对每个模型进行了五次 NT 测量，将其分为生理模型和异常模型。
将模型分类为生理 NT 或异常 NT，检测率为 100%。正常 NT 模型和增加 NT 模型的平均测量值分别为 2.27mm（SD±0.38）和 4.165mm（SD±0.51）。通过组内相关系数（ICC）计算的组内可靠性结果为 0.883，表明评分者之间存在稳健的一致性。成本效益分析表明 3D 打印过程具有经济性。
本研究强调了 3D 打印胎儿模型通过高组内可靠性、不同专家水平的一致性以及成本效益，在增强超声培训方面的潜力。需要进一步探讨的局限性包括人群变异性和对临床结果的直接转化。该研究为将创新技术纳入医学教育的讨论做出了贡献，提供了一种实用且经济的方法来获得、完善和修正产前超声诊断技能。未来的研究应探索更广泛的应用和长期的经济影响，为医学培训和实践的变革性进步铺平道路。
