Department of Circulation and Medical Imaging, NTNU, Norwegian University of Science and Technology, 7030, Trondheim, Norway.
Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim University Hospital, 7030, Trondheim, Norway.
MAGMA. 2021 Apr;34(2):309-321. doi: 10.1007/s10334-020-00871-3. Epub 2020 Jul 31.
To develop and evaluate an automated method for prostate T2-weighted (T2W) image normalization using dual-reference (fat and muscle) tissue.
Transverse T2W images from the publicly available PROMISE12 (N = 80) and PROSTATEx (N = 202) challenge datasets, and an in-house collected dataset (N = 60) were used. Aggregate channel features object detectors were trained to detect reference fat and muscle tissue regions, which were processed and utilized to normalize the 3D images by linear scaling. Mean prostate pseudo T2 values after normalization were compared to literature values. Inter-patient histogram intersections of voxel intensities in the prostate were compared between our approach, the original images, and other commonly used normalization methods. Healthy vs. malignant tissue classification performance was compared before and after normalization.
The prostate pseudo T2 values of the three tested datasets (mean ± standard deviation = 78.49 ± 9.42, 79.69 ± 6.34 and 79.29 ± 6.30 ms) corresponded well to T2 values from literature (80 ± 34 ms). Our normalization approach resulted in significantly higher (p < 0.001) inter-patient histogram intersections (median = 0.746) than the original images (median = 0.417) and most other normalization methods. Healthy vs. malignant classification also improved significantly (p < 0.001) in peripheral (AUC 0.826 vs. 0.769) and transition (AUC 0.743 vs. 0.678) zones.
An automated dual-reference tissue normalization of T2W images could help improve the quantitative assessment of prostate cancer.
开发并评估一种使用双参考（脂肪和肌肉）组织对前列腺 T2 加权（T2W）图像进行归一化的自动方法。
使用了来自公开的 PROMISE12（N=80）和 PROSTATEx（N=202）挑战赛数据集以及内部收集的数据集（N=60）的横向 T2W 图像。聚合通道特征目标检测器被训练来检测参考脂肪和肌肉组织区域，这些区域经过处理并用于通过线性缩放来归一化 3D 图像。归一化后前列腺的平均伪 T2 值与文献值进行了比较。比较了我们的方法、原始图像和其他常用归一化方法在前列腺中体素强度的患者间直方图交点。比较了归一化前后健康与恶性组织的分类性能。
三个测试数据集（平均±标准差=78.49±9.42、79.69±6.34 和 79.29±6.30 ms）的前列腺伪 T2 值与文献中的 T2 值（80±34 ms）非常吻合。与原始图像（中位数=0.417）和大多数其他归一化方法相比，我们的归一化方法导致的患者间直方图交点（中位数=0.746）显著更高（p<0.001）。外周（AUC 0.826 与 0.769）和过渡（AUC 0.743 与 0.678）区的健康与恶性分类也显著改善（p<0.001）。
T2W 图像的自动双参考组织归一化可以帮助改善前列腺癌的定量评估。
