A comparative study between deep learning and radiomics models in grading liver tumors using hepatobiliary phase contrast-enhanced MR images

BMC Medical Imaging

Research

Jianpeng Yuan¹  Zujun Hou²  Cong Wang³  Meng Gan³  Lixin Du⁴  Zhigang Li⁴  Pan Wang⁴ 

[1] Department of Radiology, The Seventh Affiliated Hospital, Sun Yat-sen University, Shenzhen, China;Jiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China;Jiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou, China;Jinan Guoke Medical Technology Development Co., Ltd, Jinan, China;Medical Imaging Department, Shenzhen Longhua District Central Hospital, Shenzhen, China;

关键词: Hepatocellular carcinoma;    Radiomics;    Deep learning;    Magnetic resonance imaging;   

DOI  :  10.1186/s12880-022-00946-8

 received in 2022-09-06, accepted in 2022-12-02,  发布年份 2022

来源: Springer

【 摘 要 】

PurposeTo compare a deep learning model with a radiomics model in differentiating high-grade (LR-3, LR-4, LR-5) liver imaging reporting and data system (LI-RADS) liver tumors from low-grade (LR-1, LR-2) LI-RADS tumors based on the contrast-enhanced magnetic resonance images.MethodsMagnetic resonance imaging scans of 361 suspected hepatocellular carcinoma patients were retrospectively reviewed. Lesion volume segmentation was manually performed by two radiologists, resulting in 426 lesions from the training set and 83 lesions from the test set. The radiomics model was constructed using a support vector machine (SVM) with pre-defined features, which was first selected using Chi-square test, followed by refining using binary least absolute shrinkage and selection operator (LASSO) regression. The deep learning model was established based on the DenseNet. Performance of the models was quantified by area under the receiver-operating characteristic curve (AUC), accuracy, sensitivity, specificity and F1-score.ResultsA set of 8 most informative features was selected from 1049 features to train the SVM classifier. The AUCs of the radiomics model were 0.857 (95% confidence interval [CI] 0.816–0.888) for the training set and 0.879 (95% CI 0.779–0.935) for the test set. The deep learning method achieved AUCs of 0.838 (95% CI 0.799–0.871) for the training set and 0.717 (95% CI 0.601–0.814) for the test set. The performance difference between these two models was assessed by t-test, which showed the results in both training and test sets were statistically significant.ConclusionThe deep learning based model can be trained end-to-end with little extra domain knowledge, while the radiomics model requires complex feature selection. However, this process makes the radiomics model achieve better performance in this study with smaller computational cost and more potential on model interpretability.

【 授权许可】

CC BY   
© The Author(s) 2022

【 图 表 】

12982_2022_119_Article_IEq198.gif

12982_2022_119_Article_IEq197.gif

12982_2022_119_Article_IEq196.gif

12982_2022_119_Article_IEq195.gif

Fig. 2

Fig. 1

12982_2022_119_Article_IEq192.gif

12888_2022_4392_Article_IEq3.gif

12888_2022_4392_Article_IEq2.gif

12888_2022_4392_Article_IEq1.gif

12982_2022_119_Article_IEq188.gif

12982_2022_119_Article_IEq50.gif

Fig. 2

12982_2022_119_Article_IEq187.gif

12982_2022_119_Article_IEq186.gif

Fig. 1

12982_2022_119_Article_IEq184.gif

12982_2022_119_Article_IEq183.gif

12982_2022_119_Article_IEq48.gif

12982_2022_119_Article_IEq47.gif

12982_2022_119_Article_IEq42.gif

Fig. 1

12982_2022_119_Article_IEq29.gif

12982_2022_119_Article_IEq9.gif