School of Computer Science and Engineering, The Hebrew University of Jerusalem, Israel, Edmond J. Safra Campus, Givat Ram, 9190401, Jerusalem, Israel.
Department of Orthopedics, Hadassah University Medical Center, Jerusalem, Israel.
Int J Comput Assist Radiol Surg. 2023 Dec;18(12):2179-2189. doi: 10.1007/s11548-023-02907-0. Epub 2023 Apr 25.
Radiographic parameters (RPs) provide objective support for effective decision making in determining clinical treatment of distal radius fractures (DRFs). This paper presents a novel automatic RP computation pipeline for computing the six anatomical RPs associated with DRFs in anteroposterior (AP) and lateral (LAT) forearm radiographs.
The pipeline consists of: (1) segmentation of the distal radius and ulna bones with six 2D Dynamic U-Net deep learning models; (2) landmark points detection and distal radius axis computation from the segmentations with geometric methods; (3) RP computation and generation of a quantitative DRF report and composite AP and LAT radiograph images. This hybrid approach combines the advantages of deep learning and model-based methods.
The pipeline was evaluated on 90 AP and 93 LAT radiographs for which ground truth distal radius and ulna segmentations and RP landmarks were manually obtained by expert clinicians. It achieves an accuracy of 94 and 86% on the AP and LAT RPs, within the observer variability, and an RP measurement difference of 1.4 ± 1.2° for the radial angle, 0.5 ± 0.6 mm for the radial length, 0.9 ± 0.7 mm for the radial shift, 0.7 ± 0.5 mm for the ulnar variance, 2.9 ± 3.3° for the palmar tilt and 1.2 ± 1.0 mm for the dorsal shift.
Our pipeline is the first fully automatic method that accurately and robustly computes the RPs for a wide variety of clinical forearm radiographs from different sources, hand orientations, with and without cast. The computed accurate and reliable RF measurements may support fracture severity assessment and clinical management.
影像学参数（RPs）为有效决策提供客观支持，有助于确定桡骨远端骨折（DRF）的临床治疗方案。本文提出了一种新的自动 RP 计算流程，用于计算前后位（AP）和侧位（LAT）前臂 X 光片中与 DRF 相关的六个解剖学 RP。
该流程包括：（1）使用六个二维动态 U-Net 深度学习模型对桡骨和尺骨进行分割；（2）使用几何方法从分割结果中检测标志点并计算桡骨远端轴线；（3）计算 RP 并生成定量 DRF 报告和复合 AP 和 LAT 射线照片图像。这种混合方法结合了深度学习和基于模型方法的优势。
该流程在 90 张 AP 和 93 张 LAT X 光片上进行了评估，这些 X 光片的桡骨和尺骨分割以及 RP 标志点均由专家临床医生手动获得。在观察者可变性范围内，该流程在 AP 和 LAT RP 上的准确率分别达到 94%和 86%，桡骨角的 RP 测量差值为 1.4±1.2°，桡骨长度为 0.5±0.6mm，桡骨移位为 0.9±0.7mm，尺侧变异为 0.7±0.5mm，掌倾角为 2.9±3.3°，背侧移位为 1.2±1.0mm。
我们的流程是第一个能够从不同来源、不同手位、有或没有石膏的各种临床前臂 X 光片中准确、稳健地计算 RP 的全自动方法。计算出的准确可靠的 RF 测量值可能有助于骨折严重程度评估和临床管理。
