Tian Isaac Y, Liu Jason, Wong Michael C, Kelly Nisa N, Liu Yong E, Garber Andrea K, Heymsfield Steven B, Curless Brian, Shepherd John A
Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA.
University of Hawaii Cancer Center, University of Hawaii - Manoa, Honolulu, HI, USA.
NPJ Digit Med. 2025 Feb 2;8(1):79. doi: 10.1038/s41746-025-01469-6.
Body composition prediction from 3D optical imagery has previously been studied with linear algorithms. In this study, we present a novel application of deep 3D convolutional graph networks and nonlinear Gaussian process regression for human body shape parameterization and body composition estimation. We trained and tested linear and nonlinear models with ablation studies on a novel ensemble body shape dataset containing 4286 scans. Nonlinear GPR produced up to a 20% reduction in prediction error and up to a 30% increase in precision over linear regression for both sexes in 10 tested body composition variables. Deep shape features produced 6-8% reduction in prediction error over linear PCA features for males only, and a 4-14% reduction in precision error for both sexes. All coefficients of determination (R) for all predicted variables were above 0.86 and achieved lower estimation RMSEs than all previous work on 10 metrics of body composition.
此前已使用线性算法对基于3D光学图像的身体成分预测进行了研究。在本研究中，我们提出了一种深度3D卷积图网络和非线性高斯过程回归在人体形状参数化和身体成分估计方面的新应用。我们在一个包含4286次扫描的新型整体身体形状数据集上进行了消融研究，对线性和非线性模型进行了训练和测试。在10个测试的身体成分变量中，非线性高斯过程回归在预测误差方面最多降低了20%，在精度方面比线性回归提高了30%。深度形状特征仅对男性而言，预测误差比线性主成分分析特征降低了6-8%，对两性而言，精度误差降低了4-14%。所有预测变量的决定系数（R）均高于0.86，并且在10个身体成分指标上实现了比以往所有工作更低的估计均方根误差。
