Raith Stefan, Deitermann Matthias, Pankert Tobias, Li Jianzhang, Modabber Ali, Hölzle Frank, Hildebrand Frank, Eschweiler Jörg
Department of Oral and Maxillofacial Surgery, University Hospital RWTH Aachen, Aachen, Germany.
Inzipio GmbH, Aachen, Germany.
Phys Med Biol. 2025 Feb 17;70(5). doi: 10.1088/1361-6560/adabae.
The purpose of this study was to develop a robust deep learning approach trained with a smallMRI dataset for multi-label segmentation of all eight carpal bones for therapy planning and wrist dynamic analysis.A small dataset of 15 3.0-T MRI scans from five health subjects was employed within this study. The MRI data was variable with respect to the field of view (FOV), wide range of image intensity, and joint pose. Asegmentation pipeline using modified 3D U-Net was proposed. In the, a novel architecture, introduced as expansion transfer learning (ETL), cascades the use of a focused region of interest (ROI) cropped around ground truth for pretraining and a subsequent transfer by an expansion to the original FOV for a primary prediction. The bounding box around the ROI generated was utilized in thefor high-accuracy, labeled segmentations of eight carpal bones. Different metrics including dice similarity coefficient (DSC), average surface distance (ASD) and hausdorff distance (HD) were used to evaluate performance between proposed and four state-of-the-art approaches.With an average DSC of 87.8 %, an ASD of 0.46 mm, an average HD of 2.42 mm in all datasets (96.1 %, 0.16 mm, 1.38 mm in 12 datasets after exclusion criteria, respectively), the proposed approach showed an overall strongest performance than comparisons.To our best knowledge, this is the first CNN-based multi-label segmentation approach for MRI human carpal bones. The ETL introduced in this work improved the ability to localize a small ROI in a large FOV. Overall, the interplay of aapproach and ETL culminated in convincingly accurate segmentation scores despite a very small amount of image data.
本研究的目的是开发一种强大的深度学习方法，该方法使用小型MRI数据集进行训练，用于对所有八块腕骨进行多标签分割，以辅助治疗规划和手腕动态分析。本研究使用了来自五名健康受试者的15例3.0-T MRI扫描的小型数据集。MRI数据在视野（FOV）、图像强度范围和关节姿势方面存在差异。提出了一种使用改进的3D U-Net的分割管道。其中，一种名为扩展迁移学习（ETL）的新颖架构，先对围绕真实值裁剪的感兴趣区域（ROI）进行预训练，然后通过扩展到原始FOV进行后续迁移以进行初步预测。生成的ROI周围的边界框用于对八块腕骨进行高精度的标记分割。使用包括骰子相似系数（DSC）、平均表面距离（ASD）和豪斯多夫距离（HD）在内的不同指标来评估所提出的方法与四种现有最先进方法之间的性能。在所提出的方法中，所有数据集中的平均DSC为87.8%，ASD为0.46毫米，平均HD为2.42毫米（在排除标准后的12个数据集中分别为96.1%、0.16毫米、1.38毫米），与其他方法相比，该方法总体表现最强。据我们所知，这是第一种基于CNN的用于MRI人体腕骨的多标签分割方法。本工作中引入的ETL提高了在大FOV中定位小ROI的能力。总体而言，尽管图像数据量非常少，但该方法与ETL的相互作用最终产生了令人信服的准确分割分数。
