Le Elizabeth P V, Wong Mark Y Z, Rundo Leonardo, Tarkin Jason M, Evans Nicholas R, Weir-McCall Jonathan R, Chowdhury Mohammed M, Coughlin Patrick A, Pavey Holly, Zaccagna Fulvio, Wall Chris, Sriranjan Rouchelle, Corovic Andrej, Huang Yuan, Warburton Elizabeth A, Sala Evis, Roberts Michael, Schönlieb Carola-Bibiane, Rudd James H F
Department of Medicine, University of Cambridge, United Kingdom.
Department of Radiology, University of Cambridge, United Kingdom.
Eur J Radiol Open. 2024 Aug 31;13:100594. doi: 10.1016/j.ejro.2024.100594. eCollection 2024 Dec.
To assess radiomics and deep learning (DL) methods in identifying symptomatic Carotid Artery Disease (CAD) from carotid CT angiography (CTA) images. We further compare the performance of these novel methods to the conventional calcium score.
Carotid CT angiography (CTA) images from symptomatic patients (ischaemic stroke/transient ischaemic attack within the last 3 months) and asymptomatic patients were analysed. Carotid arteries were classified into culprit, non-culprit and asymptomatic. The calcium score was assessed using the Agatston method. 93 radiomic features were extracted from regions-of-interest drawn on 14 consecutive CTA slices. For DL, convolutional neural networks (CNNs) with and without transfer learning were trained directly on CTA slices. Predictive performance was assessed over 5-fold cross validated AUC scores. SHAP and GRAD-CAM algorithms were used for explainability.
132 carotid arteries were analysed (41 culprit, 41 non-culprit, and 50 asymptomatic). For asymptomatic vs symptomatic arteries, radiomics attained a mean AUC of 0.96(± 0.02), followed by DL 0.86(± 0.06) and then calcium 0.79(± 0.08). For culprit vs non-culprit arteries, radiomics achieved a mean AUC of 0.75(± 0.09), followed by DL 0.67(± 0.10) and then calcium 0.60(± 0.02). For multi-class classification, the mean AUCs were 0.95(± 0.07), 0.79(± 0.05), and 0.71(± 0.07) for radiomics, DL and calcium, respectively. Explainability revealed consistent patterns in the most important radiomic features.
Our study highlights the potential of novel image analysis techniques in extracting quantitative information beyond calcification in the identification of CAD. Though further work is required, the transition of these novel techniques into clinical practice may eventually facilitate better stroke risk stratification.
评估放射组学和深度学习（DL）方法在从颈动脉CT血管造影（CTA）图像中识别症状性颈动脉疾病（CAD）方面的应用。我们进一步将这些新方法的性能与传统的钙化评分进行比较。
分析有症状患者（过去3个月内发生缺血性中风/短暂性脑缺血发作）和无症状患者的颈动脉CT血管造影（CTA）图像。将颈动脉分为责任血管、非责任血管和无症状血管。使用阿加斯顿方法评估钙化评分。从连续14个CTA切片上绘制的感兴趣区域中提取93个放射组学特征。对于深度学习，带有和不带有迁移学习的卷积神经网络（CNN）直接在CTA切片上进行训练。通过5折交叉验证的AUC分数评估预测性能。使用SHAP和GRAD-CAM算法进行可解释性分析。
分析了132条颈动脉（41条责任血管、41条非责任血管和50条无症状血管）。对于无症状血管与有症状血管，放射组学的平均AUC为0.96（±0.02），其次是深度学习为0.86（±0.06），然后是钙化评分为0.79（±0.08）。对于责任血管与非责任血管，放射组学的平均AUC为0.75（±0.09），其次是深度学习为0.67（±0.10），然后是钙化评分为0.60（±0.02）。对于多类分类，放射组学、深度学习和钙化评分的平均AUC分别为0.95（±0.07）、0.79（±0.05）和0.71（±0.07）。可解释性分析揭示了最重要的放射组学特征中的一致模式。
我们的研究突出了新型图像分析技术在识别CAD时提取钙化以外的定量信息方面的潜力。尽管还需要进一步的工作，但将这些新技术转化为临床实践最终可能有助于更好地进行中风风险分层。
