基于深度学习的后处理CT去噪用于心肌延迟增强
本研究回顾性分析2020年7月至2021年10月行心肌延迟增强(MDE)CT检查的患者,以三次采集平均图像为金标准训练残差密集网络(RDN)进行后处理去噪,并以晚期钆增强(LGE)MRI为参考标准评估诊断性能。共纳入180例患者,RDN将图像噪声降至原水平的28%且保持CT值等效(P < .001),去噪后图像诊断敏感度、特异度和准确度均提高。
要点
- RDN去噪将MDE CT图像噪声降至原水平的28%,CT值保持等效(P < .001)。
- 以LGE MRI为参考,DLoriginal图像准确度93.8%高于原始图像82.3%,OR 1.8(95% CI: 1.2, 2.9)。
- DLave图像准确度93.8%高于平均图像88.5%,OR 2.0(95% CI: 1.2, 3.5)。
核心结论
- 以平均MDE CT图像为监督的RDN去噪网络可降低图像噪声并提高心肌延迟增强CT的诊断性能。
研究设计与样本
- 回顾性研究,纳入180例患者(中位年龄66岁,IQR 53-74岁,男107例),以100例患者的预平均图像为输入、平均图像为金标准监督训练RDN,比较原始单次采集图像、标准平均图像、RDN去噪原始图像(DLoriginal)和RDN去噪平均图像(DLave)。
关键影像表现/征象
- RDN将图像噪声降至原水平的28%,同时保持CT值等效(P < .001 for all)。
- 左心室腔和心肌的CT值与图像噪声在40例患者中评估。
- 在其余40例接受参考LGE MRI的患者中评估心肌延迟增强的节段性存在。
诊断标准与关键数值
- 原始图像诊断心肌延迟增强的敏感度77.9%、特异度84.4%、准确度82.3%。
- 平均图像敏感度89.7%、特异度87.9%、准确度88.5%。
- DLoriginal图像敏感度93.1%、特异度87.5%、准确度89.3%。
- DLave图像敏感度95.1%、特异度93.1%、准确度93.8%。
- DLoriginal较原始图像准确度提高(OR 1.8 [95% CI: 1.2, 2.9],P = .011);DLave较平均图像准确度提高(OR 2.0 [95% CI: 1.2, 3.5],P = .009)。
临床应用与阅片要点
- 所提出的以平均CT图像为监督的去噪网络可降低图像噪声并提高心肌延迟增强CT的诊断性能。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
Background To improve myocardial delayed enhancement (MDE) CT, a deep learning (DL)-based post hoc denoising method supervised with averaged MDE CT data was developed. Purpose To assess the image quality of denoised MDE CT images and evaluate their diagnostic performance by using late gadolinium enhancement (LGE) MRI as a reference. Materials and methods MDE CT data obtained by averaging three acquisitions with a single breath hold 5 minutes after the contrast material injection in patients from July 2020 to October 2021 were retrospectively reviewed. Preaveraged images obtained in 100 patients as inputs and averaged images as ground truths were used to supervise a residual dense network (RDN). The original single-shot image, standard averaged image, RDN-denoised original (DLoriginal) image, and RDN-denoised averaged (DLave) image of holdout cases were compared. In 40 patients, the CT value and image noise in the left ventricular cavity and myocardium were assessed. The segmental presence of MDE in the remaining 40 patients who underwent reference LGE MRI was evaluated. The sensitivity, specificity, and accuracy of each type of CT image and the improvement in accuracy achieved with the RDN were assessed using odds ratios (ORs) estimated with the generalized estimation equation. Results Overall, 180 patients (median age, 66 years [IQR, 53-74 years]; 107 men) were included. The RDN reduced image noise to 28% of the original level while maintaining equivalence in the CT values (P < .001 for all). The sensitivity, specificity, and accuracy of the original images were 77.9%, 84.4%, and 82.3%, of the averaged images were 89.7%, 87.9%, and 88.5%, of the DLoriginal images were 93.1%, 87.5%, and 89.3%, and of the DLave images were 95.1%, 93.1%, and 93.8%, respectively. DLoriginal images showed improved accuracy compared with the original images (OR, 1.8 [95% CI: 1.2, 2.9]; P = .011) and DLave images showed improved accuracy compared with the averaged images (OR, 2.0 [95% CI: 1.2, 3.5]; P = .009). Conclusion The proposed denoising network supervised with averaged CT images reduced image noise and improved the diagnostic performance for myocardial delayed enhancement CT. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Vannier and Wang in this issue.
引用
- Nishii, Kobayashi, Tanaka, et al. Deep Learning-based Post Hoc CT Denoising for Myocardial Delayed Enhancement. Radiology. 2022. PMID: 35762889. doi:10.1148/radiol.220189
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。