融合CBAM注意力机制的人体软组织多物理场快速同步建模

Rapid synchronous multi-physics modeling of human soft tissue with the integration of CBAM attention mechanism

摘要

人体软组织的快速物理建模对手术模拟至关重要。深度学习结合有限元计算的方案解决了传统方案建模效率低下的弊端。然而,已有的研究仅关注软组织受外载时的变形建模,忽略了应力和反力等物理场对手术训练指导性意义。为此,本文提出一种新颖的基于神经网络的软组织快速多物理场建模方案。方案通过紧凑编码软组织所受载荷与机械响应之间的非线性关系实现物理场的快速预测。针对物理场之间存在数量级差异导致网络训练不收敛的问题,引入卷积块注意力模块(convolutional block attention module, CBAM)来动态平衡各个物理场之间的权重,解决大尺度物理场主导模型训练的问题。实验表明,所提方案相较于同类方案可以更加精准、高效地预测软组织受到外部载荷时的多物理场。相比于传统数值方法,以约5%的精度损失实现了上千倍的效率提升,有望为外科模拟等计算机医疗辅助技术提供便利。

Abstract

Rapid physical modeling of human soft tissue is crucial for surgical simulation. The integration of deep learning with finite element analysis addresses the inefficiencies of traditional modeling approaches. However, existing research primarily focuses on deformation modeling under external loading, neglecting the importance of other physical fields, such as stress and reaction forces, which play a critical role in guiding surgical training. To address this gap, we propose a novel neural network-based approach for rapid multi-physics modeling of soft tissues. This approach efficiently predicts physical fields by compactly encoding the nonlinear relationships between the loading conditions and mechanical responses of the soft tissue. To resolve the issue of non-convergence during network training caused by the scale differences between physical fields, we introduce the convolutional block attention module (CBAM) to dynamically balance the weights between different physical fields, thus overcoming the dominance of large-scale physical fields in model training. Extensive experiments show that the proposed method outperforms similar approaches in terms of both accuracy and efficiency in predicting multi-physics fields of soft tissue under external loads. Compared to traditional numerical methods, it achieves a several-thousand-fold improvement in efficiency with only about a 5% loss in accuracy, making it a promising tool for computer-aided medical technologies such as surgical simulation.

关键词

物理建模 / 深度学习 / 手术模拟 / 有限元模型

Key words

physical-based modeling / deep learning / surgical simulation / finite element model

参考文献

[1] CACCIANIGA G, MARIANI A, DE PARATESI C G, et al. Multi—sensory guidance and feedback for simulation—based training in robot assisted surgery: a preliminary comparison of visual, haptic, and visuo—haptic [J]. IEEE Robotics and Automation Letters, 2021, 6(2): 3801-3808.

[2] UJITOKO Y, BAN Y. Survey of pseudo—haptics: haptic feedback design and application proposals[J]. IEEE Transactions on Haptics, 2021, 14(4): 699-711.

[3] SURYAWANSHI P, GUPTA A. A mass spring model for physically—realistic haptic rendering of deformable objects[C] // 2022 International Symposium on Electrical, Electronics and Information Engineering (ISEEIE). New Jersey: IEEE, 2022: 278-281.

[4] LI Y, ZHOU X H, LI H, et al. Haptic simulation system for liver surgery based on variable virtual stiffness optimization[C] // 2021 IEEE 7th International Conference on Virtual Reality (ICVR). Foshan: IEEE, 2021: 156-160.

[5] TANG Y S, LIU S, DENG Y R, et al. An improved method for soft tissue modeling[J]. Biomedical Signal Processing and Control, 2021, 65: 102367.

[6] MOROOKA K, CHEN X, KURAZUME R, et al. Real—time nonlinear FEM with neural network for simulating soft organ model deformation[C] // Medical Image Computing and Computer—Assisted Intervention — MICCAI 2008. Heidelberg: Springer, 2008: 742-749.

[7] LIANG L, LIU M L, MARTIN C, et al. A deep learning approach to estimate stress distribution: a fast and accurate surrogate of finite—element analysis[J]. Journal of the Royal Society, Interface, 2018, 15(138): 20170844.

[8] PHELLAN R, HACHEM B, CLIN J, et al. Real—time biomechanics using the finite element method and machine learning: review and perspective[J]. Medical Physics, 2021, 48(1): 7-18.

[9] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C] // Computer Vision — ECCV 2018. Cham: Springer, 2018: 3-19.

[10] RONNEBERGER O, FISCHER P, BROX T. U—Net: convolutional networks for biomedical image segmentation[M] // Medical Image Computing and Computer—Assisted Intervention — MICCAI 2015. Cham: Springer, 2015: 234-241.

[11] MENDIZABAL A, MÁRQUEZ—NEILA P, COTIN S. Simulation of hyperelastic materials in real—time using deep learning[J]. Medical Image Analysis, 2020, 59: 101569.

[12] DESHPANDE S, LENGI EWICZ J, BORDAS S P A. Probabilistic deep learning for real—time large deformation simulations[J]. Computer Methods in Applied Mechanics and Engineering, 2022, 398: 115307.

[13] HU J, SHEN L, SUN G. Squeeze—and—excitation networks[C] // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 7132-7141.

[14] HOU Q B, ZHOU D Q, FENG J S. Coordinate attention for efficient mobile network design[C] // 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Nashville: IEEE, 2021: 13708-13717.

[15] GUO S X, CAI X J, GAO B F, et al. Tensor—mass Model based real—time simulation of vessel deformation and force feedback for the interventional surgery training system[C] // 2017 IEEE International Conference on Mechatronics and Automation (ICMA). Takamatsu: IEEE, 2017: 433-438.

[16] YE X F, MEI X K, XIAO S G. Filling model based soft tissue deformation model[C] // 2018 IEEE International Conference on Mechatronics and Automation. Changchun: IEEE, 2018: 1655-1659.

[17] NIROOMANDI S, ALFARO I, CUETO E, et al. Real—time deformable models of non—linear tissues by model reduction techniques[J]. Computer Methods and Programs in Biomedicine, 2008, 91(3): 223-231.

[18] NIROOMANDI S, ALFARO I, CUETO E, et al. Accounting for large deformations in real—time simulations of soft tissues based on reduced—order models[J]. Computer Methods and Programs in Biomedicine, 2012, 105(1): 1-12.

[19] COURTECUISSE H, ALLARD J, KERFRIDEN P, et al. Real—time simulation of contact and cutting of heterogeneous soft—tissues[J]. Medical Image Analysis, 2014, 18(2): 394-410.

[20] LATORRE M, HUMPHREY J D. Fast, rate—independent, finite element implementation of a 3D constrained mixture model of soft tissue growth and remodeling[J]. Computer Methods in Applied Mechanics and Engineering, 2020, 368: 113156.

[21] ZHANG J N, CHAUHAN S. Fast computation of soft tissue thermal response under deformation based on fast explicit dynamics finite element algorithm for surgical simulation[J]. Computer Methods and Programs in Biomedicine, 2020, 187: 105244.

[22] WU S W, WAND T, JIANG C, et al. A finite strain model for multi—material, multi—component biomechanical analysis with total Lagrangian smoothed finite element method[J]. International Journal of Mechanical Sciences, 2023, 243: 108017.

[23] TANG L, LIU P X, HOU W G. Simulation of soft tissue deformation under physiological motion based on complementary dynamic method[J]. Computer Methods and Programs in Biomedicine, 2024, 243: 107851.

[24] BICKEL B, BÄCHER M, OTADUY M A, et al. Capture and modeling of non—linear heterogeneous soft tissue[J]. ACM Transactions on Graphics, 2009, 28(3): 1-9.

[25] PELLICER—VALERO O J, RUPÉREZ M J, MARTÍNEZ—SANCHIS S, et al. Real—time biomechanical modeling of the liver using machine learning models trained on finite element method simulations[J]. Expert Systems with Applications, 2020, 143: 113083.

[26] PFAFF T, FORTUNATO M, SANCHEZ—GONZALEZ A, et al. Learning mesh—based simulation with graph networks[C] // International Conference on Learning Representations. [S.l.: s.n.], 2021: 1-9.

[27] SALEHI Y, GIANNA COPOULOS D. PhysGNN: a physics—driven graph neural network based model for predicting soft tissue deformation in image—guided neurosurgery[J]. Advances in Neural Information Processing Systems, 2022, 35: 37282-37296.

[28] ZHANG X, WANG Z M, SUN W, et al. Heterogeneous soft tissue deformation model based on cellular neural networks: application in pulmonary hamartoma surgery[J]. Biomedical Signal Processing and Control, 2024, 95: 106290.

[29] BUSTIN H, MEYER T, REITER R, et al. ElastoNet: neural network—based multicomponent MR elastography wave inversion with uncertainty quantification[J]. Medical Image Analysis, 2025, 105: 103642.

[30] MA S Y, HE Z, WANG R K, et al. Measurement of biomechanical properties of transversely isotropic biological tissue using traveling wave expansion[J]. Medical Image Analysis, 2025, 101: 103457.

[31] KOUTRAS C, SHAYESTEHPOUR H, PÉREZ J, et al. Characterization of spine and torso stiffness via differentiable biomechanics[J]. Medical Image Analysis, 2025, 103: 103573.

[32] FANG X, KIM D, XU X A, et al. Correspondence attention for facial appearance simulation[J]. Medical Image Analysis, 2024, 93: 103094.

[33] HUANG X R, HE D M, LI Z M, et al. Maxillofacial bone movements—aware dual graph convolution approach for postoperative facial appearance prediction[J]. Medical Image Analysis, 2025, 99: 103350.

[34] CHUI C, KOBAYASHI E, CHEN X, et al. Combined compression and elongation experiments and non—linear modelling of liver tissue for surgical simulation[J]. Medical & Biological Engineering & Computing, 2004, 42(6): 787-798.

[35] UMALE S, DECK C, BOURDET N, et al. Experimental mechanical characterization of abdominal organs: liver, kidney & spleen[J]. Journal of the Mechanical Behavior of Biomedical Materials, 2013, 17: 22-33.

[36] ABDEL—MISIH S R, BLOOMSTON M. Liver anatomy[J]. Surgical Clinics of North America, 2010, 90(4): 643-653.

[37] LO B, CHUNG A J, STOYANOV D, et al. Real—time intra—operative 3D tissue deformation recovery[C] // 2008 5th IEEE International Symposium on Biomedical Imaging: From Nano to Macro. Paris: IEEE, 2008: 1387.

基金资助

国家自然科学基金资助项目(62372475)

国家自然科学基金资助项目(U22A2034)

国家自然科学基金资助项目(62177047)