Department of Electrical Engineering (ESAT), Center for Processing Speech and Images, KU Leuven, Leuven, Belgium.
UCB Pharma, Brussels, Belgium.
Osteoporos Int. 2024 Jan;35(1):143-152. doi: 10.1007/s00198-023-06903-7. Epub 2023 Sep 7.
The Convolutional Neural Network algorithm achieved a sensitivity of 94% and specificity of 93% in identifying scans with vertebral fractures (VFs). The external validation results suggest that the algorithm provides an opportunity to aid radiologists with the early identification of VFs in routine CT scans of abdomen and chest.
To evaluate the performance of a previously trained Convolutional Neural Network (CNN) model to automatically detect vertebral fractures (VFs) in CT scans in an external validation cohort.
Two Chinese studies and clinical data were used to retrospectively select CT scans of the chest, abdomen and thoracolumbar spine in men and women aged ≥50 years. The CT scans were assessed using the semiquantitative (SQ) Genant classification for prevalent VFs in a process blinded to clinical information. The performance of the CNN model was evaluated against reference standard readings by the area under the receiver operating characteristics curve (AUROC), accuracy, Cohen's kappa, sensitivity, and specificity.
A total of 4,810 subjects were included, with a median age of 62 years (IQR 56-67), of which 2,654 (55.2%) were females. The scans were acquired between January 2013 and January 2019 on 16 different CT scanners from three different manufacturers. 2,773 (57.7%) were abdominal CTs. A total of 628 scans (13.1%) had ≥1 VF (grade 2-3), representing 899 fractured vertebrae out of a total of 48,584 (1.9%) visualized vertebral bodies. The CNN's performance in identifying scans with ≥1 moderate or severe fractures achieved an AUROC of 0.94 (95% CI: 0.93-0.95), accuracy of 93% (95% CI: 93%-94%), kappa of 0.75 (95% CI: 0.72-0.77), a sensitivity of 94% (95% CI: 92-96%) and a specificity of 93% (95% CI: 93-94%).
The algorithm demonstrated excellent performance in the identification of vertebral fractures in a cohort of chest and abdominal CT scans of Chinese patients ≥50 years.
卷积神经网络算法在识别有椎体骨折（VF）的扫描中实现了 94%的灵敏度和 93%的特异性。外部验证结果表明，该算法为放射科医生在常规腹部和胸部 CT 扫描中早期识别 VF 提供了机会。
评估一种先前训练的卷积神经网络（CNN）模型在外部验证队列中自动检测 CT 扫描中椎体骨折（VF）的性能。
使用两项中国研究和临床数据，回顾性选择年龄≥50 岁的男性和女性的胸部、腹部和胸腰椎 CT 扫描。使用半定量（SQ）Genant 分类法对临床信息进行盲法评估，对 CT 扫描进行现患 VF 评估。使用受试者工作特征曲线（AUROC）下面积、准确性、Cohen's kappa、灵敏度和特异性评估 CNN 模型对参考标准读数的性能。
共纳入 4810 名受试者，中位年龄为 62 岁（IQR 56-67），其中 2654 名（55.2%）为女性。扫描时间为 2013 年 1 月至 2019 年 1 月，来自三个不同制造商的 16 台不同 CT 扫描仪。2773 例（57.7%）为腹部 CT。共有 628 例（13.1%）扫描存在≥1 个 VF（2-3 级），代表 48584 个（1.9%）可见椎体中有 899 个骨折椎体。CNN 识别≥1 个中度或重度骨折扫描的性能，AUROC 为 0.94（95%CI：0.93-0.95），准确率为 93%（95%CI：93%-94%），kappa 值为 0.75（95%CI：0.72-0.77），灵敏度为 94%（95%CI：92-96%），特异性为 93%（95%CI：93%-94%）。
该算法在识别中国≥50 岁患者的胸部和腹部 CT 扫描中的椎体骨折方面表现出优异的性能。
