Suppr超能文献

基于双能 CT 的食管癌新辅助化疗后淋巴结影像组学特征

Radiomics Features Using Dual-energy CT for Lymph Nodes After Preoperative Chemotherapy for Esophageal Cancer.

机构信息

Department of Gastroenterological Surgery and Pediatric Surgery, Graduate School of Medicine, Gifu University, Gifu, Japan;

Department of Gastroenterological Surgery and Pediatric Surgery, Graduate School of Medicine, Gifu University, Gifu, Japan.

出版信息

Anticancer Res. 2024 Apr;44(4):1661-1674. doi: 10.21873/anticanres.16965.

Abstract

BACKGROUND/AIM: Progress has been made in a triplet preoperative chemotherapy regimen for advanced esophageal cancer. We performed a preliminary investigation of the radiomics features of pathological lymph node metastasis after neoadjuvant chemotherapy using dual-energy computed tomography (DECT).

PATIENTS AND METHODS

From January to December 2022, 36 lymph nodes from 10 patients with advanced esophageal cancer who underwent contrast-enhanced DECT after neoadjuvant chemotherapy and radical surgery in our department were studied. Radiomics features were extracted from iodine-based material decomposition images at the portal venous phase constructed by DECT using MATLAB analysis software. Receiver operating characteristic (ROC) analysis and cut-off values were determined for the presence or absence of pathological metastasis.

RESULTS

ROC for the short axis of the pathologically positive lymph nodes showed an AUC of 0.713. Long run emphasis (LRE) within gray-level run-length matrix (GLRLM) was confirmed with a high AUC of 0.812. Sensitivity and specificity for lymph nodes with a short axis >10 mm were 0.222 and 1, respectively. Sensitivity and specificity for LRE within GLRLM were 0.722 and 0.833, respectively. Sensitivity and specificity for small zone emphasis (SZE) within gray-level size zone matrix (GLSZM) were 0.889 and 0.667, and zone percentage (ZP) values within GLSZM were 0.722 and 0.778, respectively. Discrimination of existing metastases using radiomics showed significantly higher sensitivity compared to lymph node short axis >10 mm (odds ratios of LRE, SZE, and ZP: 9.1, 28, and 9.1, respectively).

CONCLUSION

Evaluation of radiomics analysis using DECT may enable a more detailed evaluation of lymph node metastasis after neoadjuvant chemotherapy.

摘要

背景/目的:在三药术前化疗方案治疗晚期食管癌方面已经取得了进展。我们使用双能 CT(DECT)对新辅助化疗后病理淋巴结转移的放射组学特征进行了初步研究。

患者与方法

2022 年 1 月至 12 月,我们对我科 10 例接受新辅助化疗和根治性手术的晚期食管癌患者在门静脉期增强 DECT 后检查的 36 个淋巴结进行了研究。使用 MATLAB 分析软件从 DECT 构建的碘基物质分解图像中提取放射组学特征。为确定病理转移的存在与否,进行了 ROC 分析和截断值的确定。

结果

ROC 分析显示,病理阳性淋巴结的短轴 AUC 为 0.713。灰度游程长度矩阵(GLRLM)中的长运行强调(LRE)具有较高的 AUC(0.812)。短轴>10mm 淋巴结的灵敏度和特异性分别为 0.222 和 1。GLRLM 中的 LRE 的灵敏度和特异性分别为 0.722 和 0.833。灰度大小区域矩阵(GLSZM)中的小区域强调(SZE)的灵敏度和特异性分别为 0.889 和 0.667,GLSZM 中的区域百分比(ZP)值分别为 0.722 和 0.778。放射组学对存在转移的鉴别能力明显高于短轴>10mm 的淋巴结(LRE、SZE 和 ZP 的比值比分别为 9.1、28 和 9.1)。

结论

使用 DECT 进行放射组学分析评估可能使新辅助化疗后淋巴结转移的评估更加详细。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验