医学
阶段(地层学)
鼻咽癌
无线电技术
肿瘤科
内科学
T级
放射科
化疗
放射治疗
单变量
癌症
多元统计
机器学习
生物
计算机科学
古生物学
作者
Furong Zeng,Kairong Lin,Yabin Jin,Haojiang Li,Qiang Quan,Jian-Chun Su,Kai Chen,Jing Zhang,Chen Han,Guoyi Zhang
标识
DOI:10.1016/j.mri.2022.02.005
摘要
The purpose of this study was to explore the prognostic value of imaging features and related models in nasopharyngeal carcinoma (NPC) patients that received neoadjuvant chemotherapy.We systematically reviewed the data of 110 NPC patients who received radiotherapy and neoadjuvant chemotherapy. The patients were randomly divided into the training cohort (n = 88) and the verification cohort (n = 22). The imaging data collected in this study were screened via Pyramidics and used to construct prediction models based on histology and clinical nomographs. The models' accuracy was evaluated via calibration curves and the consistency index (C-index). In addition, we also explored the correlation between radiomics expression patterns, quantitative histological characteristics, and clinical data and then constructed a model to predict the prognosis of NPC.The models that integrated radiomics contours with all the clinical data were superior to those based on the clinical data alone (C-index 0.746 vs. C-index 0.814, respectively) and the calibration curves showed good consistency. The heat map showed that the radiomics expression pattern and selected histological characteristics were correlated with the clinical stage, T stage, and N stage (p < 0.05), and no radiomics feature was associated with lactate dehydrogenase expression, lymphocyte count, or mononuclear cell count.MRI-based radiomics can significantly improve the efficacy of traditional TNM staging and clinical data in predicting the progression-free survival (PFS) of patients with advanced NPC, which may provide an opportunity for precision medicine.
科研通智能强力驱动
Strongly Powered by AbleSci AI