Suppr超能文献

射频消融治疗结直肠癌肺转移瘤:消融区早期放射组学分析有助于检测局部肿瘤进展吗?

Radiofrequency ablation of lung metastases of colorectal cancer: could early radiomics analysis of the ablation zone help detect local tumor progression?

机构信息

Department of Diagnostic and Interventional Oncological Imaging, Institut Bergonié, Regional Comprehensive Cancer of Nouvelle-Aquitaine, Bordeaux cedex, France.

Univ. Bordeaux, Bordeaux, France.

出版信息

Br J Radiol. 2023 Jun 1;96(1146):20201371. doi: 10.1259/bjr.20201371. Epub 2023 Apr 17.

Abstract

OBJECTIVES

To determine whether radiomics data can predict local tumor progression (LTP) following radiofrequency ablation (RFA) of colorectal cancer (CRC) lung metastases on the first revaluation chest CT.

METHODS

This case-control single-center retrospective study included 95 distinct lung metastases treated by RFA (in 39 patients, median age: 63.1 years) with a contrast-enhanced CT-scan performed 3 months after RFA. Forty-eight radiomics features (RFs) were extracted from the 3D-segmentation of the ablation zone. Several supervised machine-learning algorithms were trained in 10-fold cross-validation on reproducible RFs to predict LTP, with/without denoising CT-scans. An unsupervised classification based on reproducible RFs was built with k-means algorithm.

RESULTS

There were 20/95 (26.7%) relapses within a median delay of 10 months. The best model was a stepwise logistic regression on raw CT-scans. Its cross-validated performances were: AUROC = 0.72 (0.58-0.86), area under the Precision-Recall curve (AUPRC) = 0.44. Cross-validated balanced-accuracy, sensitivity and specificity were 0.59, 0.25 and 0.93, respectively, using = 0.5 to dichotomize the model predicted probabilities ( 0.71, 0.70 and 0.72, respectively using = 0.188 according to Youden index). The unsupervised approach identified two clusters, which were not associated with LTP ( = 0.8211) but with the occurrence of per-RFA intra-alveolar hemorrhage, post-RFA cavitations and fistulizations ( = 0.0150).

CONCLUSION

Predictive models using RFs from the post-RFA ablation zone on the first revaluation CT-scan of CRC lung metastases seemed moderately informative regarding the occurrence of LTP.

ADVANCES IN KNOWLEDGE

Radiomics approach on interventional radiology data is feasible. However, patterns of heterogeneity detected with RFs on early re-evaluation CT-scans seem biased by different healing processes following benign RFA complications.

摘要

目的

在射频消融(RFA)治疗结直肠癌(CRC)肺转移后的首次胸部 CT 复查时,确定基于影像组学数据能否预测局部肿瘤进展(LTP)。

方法

本病例对照的单中心回顾性研究纳入了 95 个经 RFA 治疗的不同肺转移灶(在 39 名患者中,中位年龄:63.1 岁),在 RFA 后 3 个月进行了增强 CT 扫描。从消融区域的 3D 分割中提取了 48 个影像组学特征(RFs)。在可重复的 RFs 上使用 10 折交叉验证训练了几种有监督的机器学习算法,以预测有无降噪 CT 扫描的 LTP。使用 k-means 算法构建基于可重复 RFs 的无监督分类。

结果

95 例中有 20 例(26.7%)在中位 10 个月的时间内复发。最佳模型是基于原始 CT 扫描的逐步逻辑回归。其交叉验证性能为:AUROC=0.72(0.58-0.86),精准度-召回曲线下面积(AUPRC)=0.44。使用 =0.5 将模型预测概率二分,交叉验证的平衡准确率、敏感度和特异度分别为 0.59、0.25 和 0.93(使用 Youden 指数根据 =0.188 二分时分别为 0.71、0.70 和 0.72)。无监督方法识别出两个聚类,这些聚类与 LTP 无关( =0.8211),但与 RFA 后肺泡内出血、RFA 后空洞和瘘管形成的发生有关( =0.0150)。

结论

在结直肠癌肺转移患者 RFA 后首次 CT 复查时,使用基于 RFs 的预测模型对于 LTP 的发生具有中等程度的信息价值。

知识进展

在介入放射学数据中使用影像组学方法是可行的。然而,在早期复查 CT 扫描上通过 RFs 检测到的异质性模式似乎受到良性 RFA 并发症后不同愈合过程的影响。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/5c6c/10230393/d9b57d094d75/bjr.20201371.g001.jpg

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