Differentiating Intracranial Solitary Fibrous Tumors from Atypical Meningiomas Using Conventional MRI: A Comparison of Radiomics Models Versus Assessment from Radiologists, and Exploration of Correlation with Collagen Volume Fraction.

Author:

Zhang Hua1,Li Xiaoling2,Xing Zhen1,Kang Jie1,Song Yang3,Li JianYe4,Cao Dairong1,Wang XingFu1

Affiliation:

1. First Affiliated Hospital of Fujian Medical University

2. The Second Hospital of Longyan

3. Siemens Healthineers Ltd

4. Gutian County Hospital

Abstract

Abstract Purpose: To investigate the performance of qualitative, quantitative assessment and radiomics models between solitary fibrous tumors (SFTs) and atypical meningiomas (ATMs), and the relationship between imaging parameters and collagen volume fraction (CVF). Methods: This study included 132 patients with ATMs and 50 patients with SFTs who underwent MRI. Qualitative assessment was performed using a five-point-scale visual scoring on T1WI and T2WI. Quantitative assessment was conducted by dividing the signal values of tumors by those of centrum semioval or thalamus. Radiomics models were established on T1WI, T2WI, and T1&T2WI. Spearman correlation was used to determine the correlation between CVF and imaging features. Results The AUCs and cutoff values of qualitative parameters were 0.913, > 3 on T1WI, and 0.763, ≤ 4 on T2WI in testing set. The AUCs were 0.854, 0.877, 0.942 for T1WI, T2WI and T1&T2WI models in testing set. There were significant differences between radiomics models and qualitative and quantitative assessment in training set (P < 0.05), while there were no significant differences in testing set. Among SFTs, negative correlation was obtained between CVF and qualitative and quantitative parameters on T1WI and T2WI. Conclusion The hyperintensity on T1WI and hypointensity on T2WI were valuable for discriminating SFTs and ATMs. The performance of radiomics models was superior to qualitative and quantitative parameters. The imaging parameters on T1WI and T2WI were negatively correlated with CVF among SFTs.

Publisher

Research Square Platform LLC

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3