CTLESS: A scatter-window projection and deep learning-based transmission-less attenuation compensation method for myocardial perfusion SPECT

arXiv - PHYS - Medical Physics Pub Date : 2024-09-12 DOI:arxiv-2409.07761

Zitong Yu, Md Ashequr Rahman, Craig K. Abbey, Richard Laforest, Nancy A. Obuchowski, Barry A. Siegel, Abhinav K. Jha

{"title":"CTLESS: A scatter-window projection and deep learning-based transmission-less attenuation compensation method for myocardial perfusion SPECT","authors":"Zitong Yu, Md Ashequr Rahman, Craig K. Abbey, Richard Laforest, Nancy A. Obuchowski, Barry A. Siegel, Abhinav K. Jha","doi":"arxiv-2409.07761","DOIUrl":null,"url":null,"abstract":"Attenuation compensation (AC), while being beneficial for\nvisual-interpretation tasks in myocardial perfusion imaging (MPI) by SPECT,\ntypically requires the availability of a separate X-ray CT component, leading\nto additional radiation dose, higher costs, and potentially inaccurate\ndiagnosis due to SPECT/CT misalignment. To address these issues, we developed a\nmethod for cardiac SPECT AC using deep learning and emission scatter-window\nphotons without a separate transmission scan (CTLESS). In this method, an\nestimated attenuation map reconstructed from scatter-energy window projections\nis segmented into different regions using a multi-channel input multi-decoder\nnetwork trained on CT scans. Pre-defined attenuation coefficients are assigned\nto these regions, yielding the attenuation map used for AC. We objectively\nevaluated this method in a retrospective study with anonymized clinical\nSPECT/CT stress MPI images on the clinical task of detecting defects with an\nanthropomorphic model observer. CTLESS yielded statistically non-inferior\nperformance compared to a CT-based AC (CTAC) method and significantly\noutperformed a non-AC (NAC) method on this clinical task. Similar results were\nobserved in stratified analyses with different sexes, defect extents and\nseverities. The method was observed to generalize across two SPECT scanners,\neach with a different camera. In addition, CTLESS yielded similar performance\nas CTAC and outperformed NAC method on the metrics of root mean squared error\nand structural similarity index measure. Moreover, as we reduced the training\ndataset size, CTLESS yielded relatively stable AUC values and generally\noutperformed another DL-based AC method that directly estimated the attenuation\ncoefficient within each voxel. These results demonstrate the capability of the\nCTLESS method for transmission-less AC in SPECT and motivate further clinical\nevaluation.","PeriodicalId":501378,"journal":{"name":"arXiv - PHYS - Medical Physics","volume":"1 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - PHYS - Medical Physics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07761","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Attenuation compensation (AC), while being beneficial for visual-interpretation tasks in myocardial perfusion imaging (MPI) by SPECT, typically requires the availability of a separate X-ray CT component, leading to additional radiation dose, higher costs, and potentially inaccurate diagnosis due to SPECT/CT misalignment. To address these issues, we developed a method for cardiac SPECT AC using deep learning and emission scatter-window photons without a separate transmission scan (CTLESS). In this method, an estimated attenuation map reconstructed from scatter-energy window projections is segmented into different regions using a multi-channel input multi-decoder network trained on CT scans. Pre-defined attenuation coefficients are assigned to these regions, yielding the attenuation map used for AC. We objectively evaluated this method in a retrospective study with anonymized clinical SPECT/CT stress MPI images on the clinical task of detecting defects with an anthropomorphic model observer. CTLESS yielded statistically non-inferior performance compared to a CT-based AC (CTAC) method and significantly outperformed a non-AC (NAC) method on this clinical task. Similar results were observed in stratified analyses with different sexes, defect extents and severities. The method was observed to generalize across two SPECT scanners, each with a different camera. In addition, CTLESS yielded similar performance as CTAC and outperformed NAC method on the metrics of root mean squared error and structural similarity index measure. Moreover, as we reduced the training dataset size, CTLESS yielded relatively stable AUC values and generally outperformed another DL-based AC method that directly estimated the attenuation coefficient within each voxel. These results demonstrate the capability of the CTLESS method for transmission-less AC in SPECT and motivate further clinical evaluation.

查看原文本刊更多论文

CTLESS：基于散射窗投影和深度学习的心肌灌注 SPECT 无传输衰减补偿方法

衰减补偿（AC）虽然有利于通过 SPECT 进行心肌灌注成像（MPI）的视觉解读任务，但通常需要使用单独的 X 射线 CT 组件，从而导致额外的辐射剂量、更高的成本，并可能因 SPECT/CT 错位而造成诊断不准确。为了解决这些问题，我们开发了一种使用深度学习和发射散射窗光子的心脏 SPECT AC 方法，无需单独的透射扫描（CTLESS）。在这种方法中，利用在 CT 扫描上训练的多通道输入多解码网络，将从散射能量窗投影重建的估计衰减图分割成不同的区域。将预先确定的衰减系数分配给这些区域，从而得到用于 AC 的衰减图。在一项回顾性研究中，我们使用匿名的临床SPECT/CT应力MPI图像，对该方法进行了客观评估，该方法的临床任务是检测具有厌形模型观察者的缺陷。与基于 CT 的 AC（CTAC）方法相比，CTLESS 在该临床任务中的表现在统计学上并不逊色，而且明显优于非 AC（NAC）方法。在对不同性别、缺陷范围和严重程度进行分层分析时，也观察到了类似的结果。据观察，该方法适用于两台 SPECT 扫描仪，每台扫描仪都配有不同的摄像头。此外，CTLESS 的性能与 CTAC 相似，在均方根误差和结构相似性指数测量指标上优于 NAC 方法。此外，随着训练数据集规模的缩小，CTLESS 的 AUC 值相对稳定，总体上优于另一种直接估计每个体素内衰减系数的基于 DL 的 AC 方法。这些结果证明了CTLESS方法在SPECT中进行无透射交流的能力，并促使我们进行进一步的临床评估。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

arXiv - PHYS - Medical Physics

自引率

0.00%

发文量