Progressive DeepSSM：面向图像到形状深度模型的训练方法
Progressive DeepSSM: Training Methodology for Image-To-Shape Deep Models
AI总结：
针对现有直接从医学图像构建SSM的深度学习模型性能不佳的问题，提出progressive DeepSSM多尺度训练策略，结合形状先验与深度监督损失，提升模型准确性与训练稳定性。
摘要：
统计形状建模（Statistical shape modeling, SSM）是一种在各类医学应用中研究解剖形状的实用定量工具。然而，在这些应用中直接使用3D图像仍有很长的路要走。近期的深度学习方法为减少大量预处理步骤、直接从未分割图像构建SSM铺平了道路，但这些模型的性能仍未达到预期标准。受多尺度/多分辨率学习启发，我们提出一种新的训练策略——progressive DeepSSM，用于训练图像到形状的深度学习模型。训练在多个尺度下进行，每个尺度都利用前一尺度的输出。该策略使模型能在初始尺度学习粗略形状特征，并在后续尺度逐步学习精细的细节形状特征。我们通过分割引导的多任务学习利用形状先验，并采用深度监督损失确保每个尺度的学习效果。实验从定量和定性两方面均表明，采用所提策略训练的模型性能更优。该训练方法可用于提升任何从医学图像推断解剖结构统计表示的深度学习方法的稳定性与准确性，也可被现有深度学习方法采用，以提高模型精度和训练稳定性。
英文摘要：
Statistical shape modeling (SSM) is an enabling quantitative tool to study anatomical shapes in various medical applications. However, directly using 3D images in these applications still has a long way to go. Recent deep learning methods have paved the way for reducing the substantial preprocessing steps to construct SSMs directly from unsegmented images. Nevertheless, the performance of these models is not up to the mark. Inspired by multiscale/multiresolution learning, we propose a new training strategy, progressive DeepSSM, to train image-to-shape deep learning models. The training is performed in multiple scales, and each scale utilizes the output from the previous scale. This strategy enables the model to learn coarse shape features in the first scales and gradually learn detailed fine shape features in the later scales. We leverage shape priors via segmentation-guided multi-task learning and employ deep supervision loss to ensure learning at each scale. Experiments show the superiority of models trained by the proposed strategy from both quantitative and qualitative perspectives. This training methodology can be employed to improve the stability and accuracy of any deep learning method for inferring statistical representations of anatomies from medical images and can be adopted by existing deep learning methods to improve model accuracy and training stability.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
