Medical Imaging Center, Departments of Radiology, Nuclear Medicine and Molecular Imaging, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands.
Department of Radiology, Netherlands Cancer Institute-Antoni Van Leeuwenhoek Hospital (NCI-AVL), Amsterdam, The Netherlands.
J Magn Reson Imaging. 2024 May;59(5):1800-1806. doi: 10.1002/jmri.28935. Epub 2023 Aug 12.
Single center MRI radiomics models are sensitive to data heterogeneity, limiting the diagnostic capabilities of current prostate cancer (PCa) radiomics models.
To study the impact of image resampling on the diagnostic performance of radiomics in a multicenter prostate MRI setting.
Retrospective.
Nine hundred thirty patients (nine centers, two vendors) with 737 eligible PCa lesions, randomly split into training (70%, N = 500), validation (10%, N = 89), and a held-out test set (20%, N = 148).
FIELD STRENGTH/SEQUENCE: 1.5T and 3T scanners/T2-weighted imaging (T2W), diffusion-weighted imaging (DWI), and apparent diffusion coefficient maps.
A total of 48 normalized radiomics datasets were created using various resampling methods, including different target resolutions (T2W: 0.35, 0.5, and 0.8 mm; DWI: 1.37, 2, and 2.5 mm), dimensionalities (2D/3D) and interpolation techniques (nearest neighbor, linear, Bspline and Blackman windowed-sinc). Each of the datasets was used to train a radiomics model to detect clinically relevant PCa (International Society of Urological Pathology grade ≥ 2). Baseline models were constructed using 2D and 3D datasets without image resampling. The resampling configurations with highest validation performance were evaluated in the test dataset and compared to the baseline models.
Area under the curve (AUC), DeLong test. The significance level used was 0.05.
The best 2D resampling model (T2W: Bspline and 0.5 mm resolution, DWI: nearest neighbor and 2 mm resolution) significantly outperformed the 2D baseline (AUC: 0.77 vs. 0.64). The best 3D resampling model (T2W: linear and 0.8 mm resolution, DWI: nearest neighbor and 2.5 mm resolution) significantly outperformed the 3D baseline (AUC: 0.79 vs. 0.67).
Image resampling has a significant effect on the performance of multicenter radiomics artificial intelligence in prostate MRI. The recommended 2D resampling configuration is isotropic resampling with T2W at 0.5 mm (Bspline interpolation) and DWI at 2 mm (nearest neighbor interpolation). For the 3D radiomics, this work recommends isotropic resampling with T2W at 0.8 mm (linear interpolation) and DWI at 2.5 mm (nearest neighbor interpolation).
3 TECHNICAL EFFICACY: Stage 2.
单中心 MRI 放射组学模型对数据异质性敏感，限制了当前前列腺癌（PCa）放射组学模型的诊断能力。
研究在多中心前列腺 MRI 环境中图像重采样对放射组学诊断性能的影响。
回顾性。
930 名患者（9 个中心，2 个供应商），737 个符合条件的 PCa 病变，随机分为训练集（70%，N=500）、验证集（10%，N=89）和保留测试集（20%，N=148）。
场强/序列：1.5T 和 3T 扫描仪/T2 加权成像（T2W）、扩散加权成像（DWI）和表观扩散系数图。
使用各种重采样方法创建了总共 48 个归一化放射组学数据集，包括不同的目标分辨率（T2W：0.35、0.5 和 0.8mm；DWI：1.37、2 和 2.5mm）、维度（2D/3D）和插值技术（最近邻、线性、B 样条和 Blackman 窗口 sinc）。每个数据集均用于训练用于检测临床相关 PCa（国际泌尿病理学会分级≥2）的放射组学模型。使用无图像重采样的 2D 和 3D 数据集构建基线模型。在测试数据集评估性能最佳的重采样配置，并与基线模型进行比较。
曲线下面积（AUC）、DeLong 检验。使用的显著性水平为 0.05。
最佳 2D 重采样模型（T2W：B 样条和 0.5mm 分辨率，DWI：最近邻和 2mm 分辨率）显著优于 2D 基线（AUC：0.77 与 0.64）。最佳 3D 重采样模型（T2W：线性和 0.8mm 分辨率，DWI：最近邻和 2.5mm 分辨率）显著优于 3D 基线（AUC：0.79 与 0.67）。
图像重采样对前列腺 MRI 多中心放射组学人工智能的性能有显著影响。推荐的 2D 重采样配置为各向同性重采样，T2W 为 0.5mm（B 样条插值），DWI 为 2mm（最近邻插值）。对于 3D 放射组学，本研究建议各向同性重采样，T2W 为 0.8mm（线性插值），DWI 为 2.5mm（最近邻插值）。
3 技术功效：2 级。
