Share：
Share this content in WeChat
X
[Chinese][PDF]2371
• Clinical Article •
MRI radiomics based on deep learning 3D super-resolution reconstruction technology for predicting the efficacy of TACE combined with molecular targeted drugs in the treatment of unresectable hepatocellular carcinoma
DONG Yaning  ZHU Jufang  MAO Ke  ZHAI Xiaoyang  DUAN Jinhui  HAN Dongming 

Cite this article as: DONG Y N, ZHU J F, MAO K, et al. MRI radiomics based on deep learning 3D super-resolution reconstruction technology for predicting the efficacy of TACE combined with molecular targeted drugs in the treatment of unresectable hepatocellular carcinoma[J]. Chin J Magn Reson Imaging, 2024, 15(12): 124-130. DOI:10.12015/issn.1674-8034.2024.12.018.


[Abstract] Objective To explore the value of MRI radiomics based on deep learning three-dimensional (3D) super-resolution reconstruction technology in predicting the efficacy of transcatheter arterial chemoembolization (TACE) combined with molecularly targeted drugs in treating unresectable hepatocellular carcinoma (HCC).Materials and Methods A retrospective analysis was conducted on data from 122 patients with primary HCC, divided into an objective response group (complete remission + partial remission, n=68) and a non-objective response group (progressive disease + stable disease, n=54) according to the modified response evaluation criteria in solid tumors (mRECIST). A 3D super-resolution reconstruction technique based on generative adversarial networks was used to double the resolution of MRI-enhanced arterial early images. The dataset was randomly divided into training and validation sets in an 8∶2 ratio. Radiomic features were extracted from volume of interests delineated on both pre- and post-reconstructed images, and subsequently, radiomic scores were calculated. Logistic regression classifiers were used to establish radiomic models for both pre- and post-reconstructed images. Multivariable logistic regression was employed to screen clinical characteristics and establish a clinical model. Model performance was evaluated using receiver operating characteristic (ROC) curves, with area under the curve (AUC) compared via DeLong's test. Decision curve analysis (DCA) was used to assess the clinical value of each model.Results Logistic regression analysis identified tumor diameter [odds ratio (OR) =1.311, 95% confidence interval (CI) =1.112-1.547, P<0.001] and arterial phase enhancement (OR=9.466, 95% CI=2.489-36.001, P<0.001) as independent predictors of treatment efficacy for HCC. The post-reconstruction radiomic model exhibited the best predictive performance, with an AUC of 0.883 (95% CI: 0.814-0.952) in the training set and 0.844 (95% CI: 0.656-0.999) in the validation set. These results surpassed those of the pre-reconstruction radiomic model, which had AUC values of 0.847 (95% CI: 0.765-0.928) and 0.753 (95% CI: 0.554-0.953), respectively, and the clinical model, with AUC values of 0.834 (95% CI: 0.754-0.914) and 0.760 (95% CI: 0.564-0.956), respectively. However, the differences in AUC among the models in both the training and validation sets are not statistically significant (P values all >0.05). DCA indicated that the post-reconstruction radiomic model had the greatest net clinical benefit in the training set above a threshold of 0.34 and in the validation set between 0.36-0.59 and above 0.71.Conclusions The application of MRI radiomics enhanced by 3D super-resolution reconstruction technology based on generative adversarial networks shows promise in predicting the efficacy of TACE combined with molecularly targeted therapy for unresectable HCC.
[Keywords] hepatocellular carcinoma；transcatheter arterial chemoembolization；magnetic resonance imaging；radiomics；deep learning

DONG Yaning   ZHU Jufang   MAO Ke   ZHAI Xiaoyang   DUAN Jinhui   HAN Dongming*  

Department of Magnetic Resonance, the First Affiliated Hospital of Xinxiang Medical University, Xinxiang453100, China

Corresponding author: HAN D M, E-mail: 625492590@qq.com

Conflicts of interest   None.

Received  2024-07-12
Accepted  2024-12-10
DOI: 10.12015/issn.1674-8034.2024.12.018
Cite this article as: DONG Y N, ZHU J F, MAO K, et al. MRI radiomics based on deep learning 3D super-resolution reconstruction technology for predicting the efficacy of TACE combined with molecular targeted drugs in the treatment of unresectable hepatocellular carcinoma[J]. Chin J Magn Reson Imaging, 2024, 15(12): 124-130. DOI:10.12015/issn.1674-8034.2024.12.018.

[1]
VOGEL A, MEYER T, SAPISOCHIN G, et al. Hepatocellular carcinoma[J]. Lancet, 2022, 400(10360): 1345-1362. DOI: 10.1016/s0140-6736(22)01200-4.
[2]
YANG F, XU G L, HUANG J T, et al. Transarterial chemoembolization combined with immune checkpoint inhibitors and tyrosine kinase inhibitors for unresectable hepatocellular carcinoma: efficacy and systemic immune response[J/OL]. Front Immunol, 2022, 13: 847601 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/35300339/. DOI: 10.3389/fimmu.2022.847601.
[3]
LENCIONI R, LLOVET J M, HAN G H, et al. Sorafenib or placebo plus TACE with doxorubicin-eluting beads for intermediate stage HCC: the SPACE trial[J]. J Hepatol, 2016, 64(5): 1090-1098. DOI: 10.1016/j.jhep.2016.01.012.
[4]
WANG D Y, LIU J C, LI T Q, et al. A VEGFR targeting peptide-drug conjugate (PDC) suppresses tumor angiogenesis in a TACE model for hepatocellular carcinoma therapy[J/OL]. Cell Death Discov, 2022, 8(1): 411 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/36202781/. DOI: 10.1038/s41420-022-01198-9.
[5]
SREEDHER G, HO M L, SMITH M, et al. Magnetic resonance imaging quality control, quality assurance and quality improvement[J]. Pediatr Radiol, 2021, 51(5): 698-708. DOI: 10.1007/s00247-021-05043-6.
[6]
MAI Z H, RAJAN J, VERHOYE M, et al. Robust edge-directed interpolation of magnetic resonance images[J]. Phys Med Biol, 2011, 56(22): 7287-7303. DOI: 10.1088/0031-9155/56/22/018.
[7]
CHEN X X, WANG X M, ZHANG K, et al. Recent advances and clinical applications of deep learning in medical image analysis[J/OL]. Med Image Anal, 2022, 79: 102444 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/35472844/. DOI: 10.1016/j.media.2022.102444.
[8]
MA C, RAO Y M, LU J W, et al. Structure-preserving image super-resolution[J]. IEEE Trans Pattern Anal Mach Intell, 2022, 44(11): 7898-7911. DOI: 10.1109/TPAMI.2021.3114428.
[9]
GAO K H, WU H. Advances in the study of MRI imaging in the progression, recurrence and prognosis of hepatocellular carcinoma[J]. Chin J Magn Reson Imag, 2023, 14(12): 181-186. DOI: 10.12015/issn.1674-8034.2023.12.033.
[10]
SUN X L, YOU Y N, ZHAO X X, et al. Prediction of the early recurrence of HCC patients after TACE surgery based on T2 weighted image texture analysis[J]. Chin J Magn Reson Imag, 2021, 12(8): 22-26, 32. DOI: 10.12015/issn.1674-8034.2021.08.005.
[11]
KONG C L, ZHAO Z W, CHEN W Y, et al. Prediction of tumor response via a pretreatment MRI radiomics-based nomogram in HCC treated with TACE[J]. Eur Radiol, 2021, 31(10): 7500-7511. DOI: 10.1007/s00330-021-07910-0.
[12]
CHEN M Z, KONG C L, QIAO E Q, et al. Multi-algorithms analysis for pre-treatment prediction of response to transarterial chemoembolization in hepatocellular carcinoma on multiphase MRI[J/OL]. Insights Imaging, 2023, 14(1): 38 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/36854872/. DOI: 10.1186/s13244-023-01380-2.
[13]
LUO J P, HUANG Z M, WANG M R, et al. Prognostic role of multiparameter MRI and radiomics in progression of advanced unresectable hepatocellular carcinoma following combined transcatheter arterial chemoembolization and lenvatinib therapy[J/OL]. BMC Gastroenterol, 2022, 22(1): 108 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/35260095/. DOI: 10.1186/s12876-022-02129-9.
[14]
WANG L J, GUO T D, WANG L, et al. Improving radiomic modeling for the identification of symptomatic carotid atherosclerotic plaques using deep learning-based 3D super-resolution CT angiography[J/OL]. Heliyon, 2024, 10(8): e29331 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/38644848/. DOI: 10.1016/j.heliyon.2024.e29331.
[15]
HOU M, ZHOU L, SUN J H. Deep-learning-based 3D super-resolution MRI radiomics model: superior predictive performance in preoperative T-staging of rectal cancer[J]. Eur Radiol, 2023, 33(1): 1-10. DOI: 10.1007/s00330-022-08952-8.
[16]
TAN W J, LIU P, LI X S, et al. Classification of COVID-19 pneumonia from chest CT images based on reconstructed super-resolution images and VGG neural network[J/OL]. Health Inf Sci Syst, 2021, 9(1): 10 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/33643612/. DOI: 10.1007/s13755-021-00140-0.
[17]
The General Office of the National Health Commission. Guidelines for diagnosis and treatment of primary liver cancer (2022 edition)[J]. Chin J Surg, 2022, 60(4): 273-309. DOI: 10.3760/cma.j.cn112139-20220217-00068.
[18]
CHANG Y, JEONG S W, YOUNG JANG J, et al. Recent updates of transarterial chemoembolilzation in hepatocellular carcinoma[J/OL]. Int J Mol Sci, 2020, 21(21): 8165 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/33142892/. DOI: 10.3390/ijms21218165.
[19]
XIA Y, RAVIKUMAR N, GREENWOOD J P, et al. Super-Resolution of Cardiac MR Cine Imaging using Conditional GANs and Unsupervised Transfer Learning[J/OL]. Med Image Anal, 2021, 71: 102037 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/33910110/. DOI: 10.1016/j.media.2021.102037.
[20]
FAN M, LIU Z H, XU M S, et al. Generative adversarial network-based super-resolution of diffusion-weighted imaging: application to tumour radiomics in breast cancer[J/OL]. NMR Biomed, 2020, 33(8): e4345 [2024-07-11]. https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/nbm.4345. DOI: 10.1002/nbm.4345.
[21]
UEDA T, OHNO Y, YAMAMOTO K, et al. Compressed sensing and deep learning reconstruction for women's pelvic MRI denoising: utility for improving image quality and examination time in routine clinical practice[J/OL]. Eur J Radiol, 2021, 134: 109430 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/33276249/. DOI: 10.1016/j.ejrad.2020.109430.
[22]
DE FARIAS E C, DI NOIA C, HAN C, et al. Impact of GAN-based lesion-focused medical image super-resolution on the robustness of radiomic features[J/OL]. Sci Rep, 2021, 11(1): 21361 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/34725417/. DOI: 10.1038/s41598-021-00898-z.
[23]
ZHAO Y, ZHANG J, WANG N, et al. Intratumoral and peritumoral radiomics based on contrast-enhanced MRI for preoperatively predicting treatment response of transarterial chemoembolization in hepatocellular carcinoma[J/OL]. BMC Cancer, 2023, 23(1): 1026 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/37875815/. DOI: 10.1186/s12885-023-11491-0.
[24]
SAVSANI E, SHAW C M, FORSBERG F, et al. Contrast-enhanced US evaluation of hepatocellular carcinoma response to chemoembolization: a prospective multicenter trial[J/OL]. Radiology, 2023, 309(1): e230727 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/37847138/. DOI: 10.1148/radiol.230727.
[25]
YAN B, BAI D S, ZHANG C, et al. Characteristics and risk differences of different tumor sizes on distant metastases of hepatocellular carcinoma: a retrospective cohort study in the SEER database[J]. Int J Surg, 2020, 80: 94-100. DOI: 10.1016/j.ijsu.2020.06.018.
[26]
OGAWA H, KAJI K, NISHIMURA N, et al. Lenvatinib prevents liver fibrosis by inhibiting hepatic stellate cell activation and sinusoidal capillarization in experimental liver fibrosis[J]. J Cell Mol Med, 2021, 25(8): 4001-4013. DOI: 10.1111/jcmm.16363.
[27]
YOSHIMITSU K. Transarterial chemoembolization using iodized oil for unresectable hepatocellular carcinoma: perspective from multistep hepatocarcinogenesis[J]. Hepat Med, 2014, 6: 89-94. DOI: 10.2147/HMER.S31440.
[28]
SUN R, LIMKIN E J, VAKALOPOULOU M, et al. A radiomics approach to assess tumour-infiltrating CD8 cells and response to anti-PD-1 or anti-PD-L1 immunotherapy: an imaging biomarker, retrospective multicohort study[J]. Lancet Oncol, 2018, 19(9): 1180-1191. DOI: 10.1016/S1470-2045(18)30413-3.
[29]
DERCLE L, LU L, SCHWARTZ L H, et al. Radiomics response signature for identification of metastatic colorectal cancer sensitive to therapies targeting EGFR pathway[J]. J Natl Cancer Inst, 2020, 112(9): 902-912. DOI: 10.1093/jnci/djaa017.
[30]
ZHAO M, WEI Y, WONG K K L. A Generative Adversarial Network technique for high-quality super-resolution reconstruction of cardiac magnetic resonance images[J]. Magn Reson Imaging, 2022, 85: 153-160. DOI: 10.1016/j.mri.2021.10.033.
[31]
KRAVCHENKO D, ISAAK A, MESROPYAN N, et al. Deep learning super-resolution reconstruction for fast and high-quality cine cardiovascular magnetic resonance[J/OL]. Eur Radiol, 2024: 108 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/39441391/. DOI: 10.1007/s00330-024-11145-0.
[32]
YOU A, KIM J K, RYU I H, et al. Application of generative adversarial networks (GAN) for ophthalmology image domains: a survey[J/OL]. Eye Vis, 2022, 9(1): 6 [2024-07-11]. https://pubmed.ncbi.nlm.nih.gov/35109930/. DOI: 10.1186/s40662-022-00277-3.

PREV Value of intratumoral and peritumoral radiomics based on DCE-MRI and DWI in predicting HER-2 status in breast cancer
NEXT Discrimination and diagnosis of between perianal-first-onset and non-perianal-first-onset fistulizing Crohn,s disease based on Van Assche score and MAGNIFI-CD score
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn