Mendez Mauro, Sundararaman Shruthi, Probyn Linda, Tyrrell Pascal N
Institute of Medical Science, University of Toronto, Toronto, Ontario, Canada.
Department of Medical Imaging, University of Toronto, Toronto, Ontario, Canada.
J Ultrasound Med. 2023 Dec;42(12):2695-2706. doi: 10.1002/jum.16332. Epub 2023 Sep 29.
This scoping review examines the emerging field of synthetic ultrasound generation using machine learning (ML) models in radiology. Nineteen studies were analyzed, revealing three primary methodological strategies: unconditional generation, conditional generation, and domain translation. Synthetic ultrasound is mainly used to augment training datasets and as training material for radiologists. Blind expert assessment and Fréchet Inception Distance are common evaluation methods. Current limitations include the need for large training datasets, manual annotations for controllable generation, and insufficient research on incorporating new domain knowledge. While generative ultrasound models show promise, further work is required for clinical implementation.
本综述探讨了放射学中使用机器学习（ML）模型生成合成超声这一新兴领域。分析了19项研究，揭示了三种主要的方法策略：无条件生成、条件生成和域翻译。合成超声主要用于扩充训练数据集以及作为放射科医生的训练材料。盲法专家评估和弗雷歇因距离是常用的评估方法。当前的局限性包括需要大型训练数据集、可控生成的人工标注以及在纳入新领域知识方面的研究不足。虽然生成式超声模型显示出前景，但临床应用还需要进一步的工作。
