Multi-sequence MRI radiomics of colorectal liver metastases: Which features are reproducible across readers?

被引:3
|
作者
van der Reijd, Denise J. [1 ,2 ]
Chupetlovska, Kalina [1 ]
van Dijk, Eleanor [1 ]
Westerink, Bram [1 ]
Monraats, Melanie A. [1 ]
Van Griethuysen, Joost J. M. [1 ]
Lambregts, Doenja M. J. [1 ,2 ]
Tissier, Renaud [3 ]
Beets-Tan, Regina G. H. [1 ,2 ,4 ]
Benson, Sean [1 ,5 ]
Maas, Monique [1 ,2 ]
机构
[1] Netherlands Canc Inst, Dept Radiol, Plesmanlaan 121, NL-1066 CX Amsterdam, Netherlands
[2] Maastricht Univ, GROW Sch Oncol & Reprod, Univ Singel 40, NL-6229 ER Maastricht, Netherlands
[3] Netherlands Canc Inst, Biostat Ctr, Plesmanlaan 121, NL-1066 CX Amsterdam, Netherlands
[4] Univ Southern Denmark, Fac Hlth Sci, Campusvej 55, DK-5203 Odense, Denmark
[5] Amsterdam Univ Med Ctr, Amsterdam Univ, Dept Cardiol, Med Ctr, Meibergdreef 9, NL-1105 AZ Amsterdam, Netherlands
关键词
Colorectal cancer; Hepatic metastasis; MRI; Radiomics; Reproducibility; Segmentation; TEST-RETEST; ADC MAPS; COEFFICIENT; STABILITY; CANCER;
D O I
10.1016/j.ejrad.2024.111346
中图分类号
R8 [特种医学]; R445 [影像诊断学];
学科分类号
1002 ; 100207 ; 1009 ;
摘要
Purpose: To assess the inter-reader reproducibility of radiomics features on multiple MRI sequences after segmentations of colorectal liver metastases (CRLM). Method: 30 CRLM (in 23 patients) were manually delineated by three readers on MRI before the start of chemotherapy on the contrast enhanced T1-weighted images (CE-T1W) in the portal venous phase, T2-weighted images (T2W) and b800 diffusion weighted images (DWI). DWI delineations were copied to the ADC-maps. 107 radiomics features were extracted per sequence. The intraclass correlation coefficient (ICC) was calculated per feature. Features were considered reproducible if ICC > 0.9. Results: 90% of CE-T1W features were reproducible with a median ICC of 0.98 (range 0.76-1.00). 81% of DWI features were robust with median ICC = 0.97 (range 0.38-1.00). The T2W features had a median ICC of 0.96 (range 0.55-0.99) and were reproducible in 80%. ADC showed the lowest number of reproducible features with 58% and median ICC = 0.91 (range 0.38-0.99) When considering the lower bound of the ICC 95% confidence intervals, 58%, 66%, 54% and 29% reached 0.9 for the CE-T1W, DWI, T2W and ADC features, respectively. The feature class with the best reproducibility differed per sequence. Conclusions: The majority of MRI radiomics features from CE-T1W, T2W, DWI and ADC in colorectal liver metastases were robust for segmentation variability between readers. The CE-T1W yielded slightly better reproducibility results compared to DWI and T2W. The ADC features seem more susceptible to reader differences compared to the other three sequences.
引用
收藏
页数:7
相关论文
共 27 条
  • [21] Acquisition repeatability of MRI radiomics features in the head and neck: a dual-3D-sequence multi-scan study
    Xue, Cindy
    Yuan, Jing
    Zhou, Yihang
    Wong, Oi Lei
    Cheung, Kin Yin
    Yu, Siu Ki
    VISUAL COMPUTING FOR INDUSTRY BIOMEDICINE AND ART, 2022, 5 (01)
  • [22] Prediction of Local Tumor Progression After Microwave Ablation in Colorectal Carcinoma Liver Metastases Patients by MRI Radiomics and Clinical Characteristics-Based Combined Model: Preliminary Results
    Shahveranova, Arzu
    Balli, Huseyin Tugsan
    Aikimbaev, Kairgeldy
    Piskin, Ferhat Can
    Sozutok, Sinan
    Yucel, Sevinc Puren
    CARDIOVASCULAR AND INTERVENTIONAL RADIOLOGY, 2023, 46 (06) : 713 - 725
  • [23] MRI-Based Radiomics Nomogram for Preoperatively Differentiating Intrahepatic Mass-Forming Cholangiocarcinoma From Resectable Colorectal Liver Metastases
    Xu, Ying
    Ye, Feng
    Li, Lu
    Yang, Yi
    Ouyang, Jingzhong
    Zhou, Yanzhao
    Wang, Sicong
    Xie, Lizhi
    Zhou, Jinxue
    Zhao, Hong
    Zhao, Xinming
    ACADEMIC RADIOLOGY, 2023, 30 (09) : 2010 - 2020
  • [24] Acquisition repeatability of MRI radiomics features in the head and neck: a dual-3D-sequence multi-scan study
    Cindy Xue
    Jing Yuan
    Yihang Zhou
    Oi Lei Wong
    Kin Yin Cheung
    Siu Ki Yu
    Visual Computing for Industry, Biomedicine, and Art, 5
  • [25] Prediction of Local Tumor Progression After Microwave Ablation in Colorectal Carcinoma Liver Metastases Patients by MRI Radiomics and Clinical Characteristics-Based Combined Model: Preliminary Results
    Arzu Shahveranova
    Huseyin Tugsan Balli
    Kairgeldy Aikimbaev
    Ferhat Can Piskin
    Sinan Sozutok
    Sevinc Puren Yucel
    CardioVascular and Interventional Radiology, 2023, 46 : 713 - 725
  • [26] Breast DCE-MRI radiomics: a robust computer-aided system based on reproducible BI-RADS features across the influence of datasets bias and segmentation methods
    Qiao, Mengyun
    Li, Chengkang
    Suo, Shiteng
    Cheng, Fang
    Hua, Jia
    Xue, Dan
    Guo, Yi
    Xu, Jianrong
    Wang, Yuanyuan
    INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY, 2020, 15 (06) : 921 - 930
  • [27] Breast DCE-MRI radiomics: a robust computer-aided system based on reproducible BI-RADS features across the influence of datasets bias and segmentation methods
    Mengyun Qiao
    Chengkang Li
    Shiteng Suo
    Fang Cheng
    Jia Hua
    Dan Xue
    Yi Guo
    Jianrong Xu
    Yuanyuan Wang
    International Journal of Computer Assisted Radiology and Surgery, 2020, 15 : 921 - 930