Goller Sophia S, Rischewski Jon F, Liebig Thomas, Ricke Jens, Siller Sebastian, Schmidt Vanessa F, Stahl Robert, Kulozik Julian, Baum Thomas, Kirschke Jan S, Foreman Sarah C, Gersing Alexandra S
Department of Radiology, University Hospital, LMU Munich, Marchioninistr. 15, 81377 Munich, Germany.
Institute for Diagnostic and Interventional Neuroradiology, University Hospital, LMU Munich, Marchioninistr. 15, 81377 Munich, Germany.
Diagnostics (Basel). 2023 Jun 20;13(12):2119. doi: 10.3390/diagnostics13122119.
Opportunistic osteoporosis screening using multidetector CT-scans (MDCT) and convolutional neural network (CNN)-derived segmentations of the spine to generate volumetric bone mineral density (vBMD) bears the potential to improve incidental osteoporotic vertebral fracture (VF) prediction. However, the performance compared to the established manual opportunistic vBMD measures remains unclear. Hence, we investigated patients with a routine MDCT of the spine who had developed a new osteoporotic incidental VF and frequency matched to patients without incidental VFs as assessed on follow-up MDCT images after 1.5 years. Automated vBMD was generated using CNN-generated segmentation masks and asynchronous calibration. Additionally, manual vBMD was sampled by two radiologists. Automated vBMD measurements in patients with incidental VFs at 1.5-years follow-up ( = 53) were significantly lower compared to patients without incidental VFs ( = 104) (83.6 ± 29.4 mg/cm vs. 102.1 ± 27.7 mg/cm, < 0.001). This comparison was not significant for manually assessed vBMD (99.2 ± 37.6 mg/cm vs. 107.9 ± 33.9 mg/cm, = 0.30). When adjusting for age and sex, both automated and manual vBMD measurements were significantly associated with incidental VFs at 1.5-year follow-up, however, the associations were stronger for automated measurements (β = -0.32; 95% confidence interval (CI): -20.10, 4.35; < 0.001) compared to manual measurements (β = -0.15; 95% CI: -11.16, 5.16; < 0.03). In conclusion, automated opportunistic measurements are feasible and can be useful for bone mineral density assessment in clinical routine.
利用多探测器计算机断层扫描（MDCT）和基于卷积神经网络（CNN）的脊柱分割来生成骨体积密度（vBMD）进行机会性骨质疏松筛查，有潜力改善偶然发生的骨质疏松性椎体骨折（VF）预测。然而，与既定的手动机会性vBMD测量方法相比，其性能仍不明确。因此，我们对接受脊柱常规MDCT检查的患者进行了研究，这些患者出现了新的偶然发生的骨质疏松性VF，并且在1.5年后的随访MDCT图像上与未发生偶然VF的患者进行了频率匹配。使用CNN生成的分割掩码和异步校准生成自动vBMD。此外，由两名放射科医生采集手动vBMD。在1.5年随访时，发生偶然VF的患者（n = 53）的自动vBMD测量值明显低于未发生偶然VF的患者（n = 104）（83.6±29.4mg/cm² 对102.1±27.7mg/cm²，P < 0.001）。对于手动评估的vBMD，这种比较无统计学意义（99.2±37.6mg/cm² 对107.9±33.9mg/cm²，P = 0.30）。在调整年龄和性别后，自动和手动vBMD测量在1.5年随访时均与偶然VF显著相关，然而，与手动测量相比，自动测量的相关性更强（β = -0.32；95%置信区间（CI）：-20.10，4.35；P < 0.001）（β = -0.15；95%CI：-11.16，5.16；P < 0.03）。总之，自动机会性测量是可行的，并且在临床常规中可用于骨密度评估。
