Zhang Hehui, Wei Wen, Qian Baoxin, Wu Daoqin, Zheng Cunhong, Li Honghua, Tang Jinsong
The Second Affiliated Hospital of Fujian University of Traditional Chinese Medicine, Fuzhou, Fujian, China.
Huiying Medical Technology Co., Ltd, Beijing City, 100192, China.
Heliyon. 2023 Oct 12;9(10):e20750. doi: 10.1016/j.heliyon.2023.e20750. eCollection 2023 Oct.
To explore the differences between low kiloelectron volt (keV) virtual monoenergetic images (VMIs) using IQon spectral CT and conventional CT (120 kVp) in the diagnosis of osteoporosis.
This retrospective study included 317 patients who underwent IQon spectral CT and dual-energy X-ray absorptiometry (DXA) examination. Commercial deep learning-based software was used for the fully automated extraction of the CT values of the first to fourth lumbar vertebrae (L1-L4) from two different low-keV levels (including 40/70 keV) VMIs and conventional 120 kVp images. The DXA examination results served as the standard of reference (normal [T-score ≥ -1], osteopenia [-2.5 < T-score < -1], and osteoporosis [T-score < -2.5]). Osteoporosis diagnosis models were constructed using machine learning classifiers (logistic regression, support vector machine, random forest, XGBoost, and multilayer perceptron) based on the average CT values of L1-L4. The area under the receiver operating characteristic curve (AUC) and DeLong test were performed to compare differences in the performance of the osteoporosis diagnosis model between virtual low-keV VMIs and standard 120 kVp images.
Random forest-based prediction model obtained good overall performance among all classifiers, and macro/micro average AUC values of 0.820/840, 0.834/853, and 0.831/852 were obtained based on 40/70 keV and 120 kVp images, respectively. The model presented no significant difference between low-keV VMIs and standard 120 kVp images for the diagnosis of osteoporosis (p > 0.05).
The performance of the osteoporosis diagnosis model using IQon spectral CT simulating the low tube voltage scanning condition (less than 120 kVp) was also satisfactory. Bone density screening evaluation can be performed with a combination of low-dose lung scanning CT, greatly reducing the radiation dose without affecting the diagnosis.
探讨使用IQon光谱CT的低千电子伏特（keV）虚拟单能量图像（VMI）与传统CT（120 kVp）在骨质疏松症诊断中的差异。
这项回顾性研究纳入了317例行IQon光谱CT和双能X线吸收法（DXA）检查的患者。使用基于深度学习的商业软件从两个不同的低keV水平（包括40/70 keV）的VMI和传统的120 kVp图像中全自动提取第一至第四腰椎（L1-L4）的CT值。DXA检查结果作为参考标准（正常[T值≥ -1]、骨量减少[-2.5 < T值< -1]和骨质疏松症[T值< -2.5]）。基于L1-L4的平均CT值，使用机器学习分类器（逻辑回归、支持向量机、随机森林、XGBoost和多层感知器）构建骨质疏松症诊断模型。进行受试者操作特征曲线（AUC）下面积和DeLong检验，以比较虚拟低keV VMI和标准120 kVp图像之间骨质疏松症诊断模型性能的差异。
基于随机森林的预测模型在所有分类器中获得了良好的总体性能，基于40/70 keV和120 kVp图像分别获得了0.820/840、0.834/853和0.831/852的宏观/微观平均AUC值。该模型在低keV VMI和标准120 kVp图像之间对骨质疏松症的诊断无显著差异（p > 0.05）。
使用IQon光谱CT模拟低管电压扫描条件（小于120 kVp）的骨质疏松症诊断模型性能也令人满意。可结合低剂量肺部扫描CT进行骨密度筛查评估，在不影响诊断的情况下大幅降低辐射剂量。
