ETS Montreal, University of Quebec, 1100 Notre-Dame West, Montreal, QC, Canada.
Spinologics Inc., 6750 Esplanade Avenue #290, Montreal, QC, Canada.
Med Phys. 2021 Jan;48(1):7-18. doi: 10.1002/mp.14602. Epub 2020 Dec 7.
The finite element method (FEM) is the preferred method to simulate phenomena in anatomical structures. However, purely FEM-based mechanical simulations require considerable time, limiting their use in clinical applications that require real-time responses, such as haptics simulators. Machine learning (ML) approaches have been proposed to help with the reduction of the required time. The present paper reviews cases where ML could help to generate faster simulations, without considerably affecting the performance results.
This review details the ML approaches used, considering the anatomical structures involved, the data collection strategies, the selected ML algorithms, with corresponding features, the metrics used for validation, and the resulting time gains.
A total of 41 references were found. ML algorithms are mainly trained with FEM-based simulations in 32 publications. The preferred ML approach is neural networks, including deep learning in 35 publications. Tissue deformation is simulated in 18 applications, but other features are also considered. The average distance error and mean squared error are the most frequently used performance metrics, in 14 and 17 publications, respectively. The time gains were considerable, going from hours or minutes for purely FEM-based simulations to milliseconds, when using ML.
ML algorithms can be used to accelerate FEM-based biomechanical simulations of anatomical structures, possibly reaching real-time responses. Fast and real-time simulations of anatomical structures, generated with ML algorithms, can help to reduce the time required by FEM-based simulations and accelerate their adoption in the clinical practice.
有限元法（FEM）是模拟解剖结构中现象的首选方法。然而，纯粹基于 FEM 的机械模拟需要相当长的时间，限制了它们在需要实时响应的临床应用中的使用，例如触觉模拟器。已经提出了机器学习（ML）方法来帮助减少所需的时间。本文回顾了 ML 有助于生成更快模拟的情况，而不会对性能结果产生显著影响。
本综述详细介绍了所使用的 ML 方法，考虑了涉及的解剖结构、数据收集策略、所选的 ML 算法、相应的特征、用于验证的指标以及由此产生的时间增益。
共发现 41 篇参考文献。在 32 篇出版物中，ML 算法主要是用基于 FEM 的模拟进行训练的。首选的 ML 方法是神经网络，包括 35 篇出版物中的深度学习。在 18 个应用中模拟了组织变形，但也考虑了其他特征。平均距离误差和均方误差是最常用的性能指标，分别在 14 篇和 17 篇出版物中使用。使用 ML 后，时间增益相当可观，从基于 FEM 的模拟的数小时或数分钟缩短到毫秒。
ML 算法可用于加速解剖结构的基于 FEM 的生物力学模拟，可能达到实时响应。使用 ML 算法生成的解剖结构的快速实时模拟可以帮助减少基于 FEM 的模拟所需的时间，并加速其在临床实践中的采用。
