A deep learning model based on magnetic resonance imaging and clinical feature fusion for predicting preoperative cytokeratin 19 status in hepatocellular carcinoma
Objective To establish a deep learning model for testing the feasibility of combining magnetic resonance imaging (MRI) deep learning features with clinical features for preoperative prediction of cytokeratin 19 (CK19) status of hepatocellular carcinoma (HCC). Methods A retrospective experiment was conducted based on the data of 116 HCC patients with confirmed CK19 status. A single sequence multi-scale feature fusion deep learning model (MSFF-IResnet) and a multi-scale and multi-modality feature fusion model (MMFF-IResnet) were established based on the hepatobiliary phase (HBP), diffusion weighted imaging (DWI) sequences of enhanced MRI images, and the clinical features significantly correlated with CK19 status. The classification performance of the models were evaluated to assess the effectiveness of the deep learning models for predicting CK19 status of HCC before surgery. Results Multivariate analysis showed that an increased neutrophil-to-lymphocyte ratio (P=0.029) and incomplete tumor capsule (P=0.028) were independent predictors of CK19 expression in HCC. The deep learning models improved by multi-scale feature fusion and multi-modality feature fusion methods achieved better classification results than the traditional machine learning models and baseline models, and the final MMFF-IResnet model showed the best classification performance with an AUC of 84.2%, an accuracy of 80.6%, a sensitivity of 80.1% and a specificity of 81.2%. Conclusion The multi-scale and multi-modality feature fusion model based on MRI and clinical parameters is capable of predicting CK19 status of HCC, demonstrating the feasibility of combining deep learning methods with MRI and clinical features for preoperative prediction of CK19 status.
WuMS, ZhongJH, ChenK, et al. Association of CK19 expression with the efficacy of adjuvant transarterial chemoembolization after hepatic resection in hepatocellular carcinoma patients at high risk of recurrence[J]. J Clin Transl Res, 2022, 8(1): 71-9.
[3]
ZhuoJY, LuD, TanWY, et al. CK19-positive hepatocellular carcinoma is a characteristic subtype[J]. J Cancer, 2020, 11(17): 5069-77.
[4]
GuoYX, ChenJJ, ZhangYF, et al. Differentiating Cytokeratin 19 expression of hepatocellular carcinoma by using multi-b-value diffusion-weighted MR imaging with mono-exponential, stretched exponential, intravoxel incoherent motion, diffusion kurtosis imaging and fractional order calculus models[J]. Eur J Radiol, 2022, 150: 110237.
ChoiSY, KimSH, ParkCK, et al. Imaging features of gadoxetic acid-enhanced and diffusion-weighted MR imaging for identifying cytokeratin 19-positive hepatocellular carcinoma: a retrospective observational study[J]. Radiology, 2018, 286(3): 897-908.
[7]
ChenYY, ChenJ, ZhangY, et al. Preoperative prediction of cytokeratin 19 expression for hepatocellular carcinoma with deep learning radiomics based on gadoxetic acid-enhanced magnetic resonance imaging[J]. J Hepatocell Carcinoma, 2021, 8: 795-808.
[8]
YonedaN, SatoY, KitaoA, et al. Epidermal growth factor induces cytokeratin 19 expression accompanied by increased growth abilities in human hepatocellular carcinoma[J]. Lab Invest, 2011, 91(2): 262-72.
[9]
YoonJK, ChoiJY, RheeH, et al. MRI features of histologic subtypes of hepatocellular carcinoma: correlation with histologic, genetic, and molecular biologic classification[J]. Eur Radiol, 2022, 32(8): 5119-33.
[10]
HuXX, WangWT, YangL, et al. MR features based on LI-RADS identify cytokeratin 19 status of hepatocellular carcinomas[J]. Eur J Radiol, 2019, 113: 7-14.
[11]
WangWT, GuDS, WeiJW, et al. A radiomics-based biomarker for cytokeratin 19 status of hepatocellular carcinoma with gadoxetic acid-enhanced MRI[J]. Eur Radiol, 2020, 30(5): 3004-14.
[12]
LuXY, XiT, LauWY, et al. Hepatocellular carcinoma expressing cholangiocyte phenotype is a novel subtype with highly aggressive behavior[J]. Ann Surg Oncol, 2011, 18(8): 2210-7.
[13]
LeeCW, LinSE, YuMC, et al. Does neutrophil to lymphocyte ratio have a role in identifying cytokeratin 19-expressing hepatocellular carcinoma?[J]. J Pers Med, 2021, 11(11): 1078.
[14]
GovaereO, KomutaM, BerkersJ, et al. Keratin 19: a key role player in the invasion of human hepatocellular carcinomas[J]. Gut, 2014, 63(4): 674-85.
[15]
SunDW, ZhangYY, SunXD, et al. Prognostic value of cytokeratin 19 in hepatocellular carcinoma: a meta-analysis[J]. Clin Chim Acta, 2015, 448: 161-9.
[16]
TsuchiyaK, KomutaM, YasuiY, et al. Expression of keratin 19 is related to high recurrence of hepatocellular carcinoma after radiofrequency ablation[J]. Oncology, 2011, 80(3/4): 278-88.
[17]
QinSD, ZhangJ, QiYP, et al. Individual and joint influence of cytokeratin 19 and microvascular invasion on the prognosis of patients with hepatocellular carcinoma after hepatectomy[J]. World J Surg Oncol, 2022, 20(1): 209.
[18]
YangF, WanYD, XuL, et al. MRI-radiomics prediction for cytokeratin 19-positive hepatocellular carcinoma: a multicenter study[J]. Front Oncol, 2021, 11: 672126.
[19]
CavigliaGP, CiruoloM, OliveroA, et al. Prognostic role of serum cytokeratin-19 fragment (CYFRA 21-1) in patients with hepatocellular carcinoma[J]. Cancers, 2020, 12(10): 2776.
[20]
ZhaoY, TanXL, ChenJM, et al. Preoperative prediction of cytokeratin-19 expression for hepatocellular carcinoma using T1 mapping on gadoxetic acid-enhanced MRI combined with diffusion-weighted imaging and clinical indicators[J]. Front Oncol, 2022, 12: 1068231.
LvJ, YinHY, YuHJ, et al. The added value of 18F-FDG PET/MRI multimodal imaging in hepatocellular carcinoma for identifying cytokeratin 19 status[J]. Abdom Radiol, 2023, 48(7): 2331-9.
HuangXL, LongLL, WeiJQ, et al. Radiomics for diagnosis of dual-phenotype hepatocellular carcinoma using Gd-EOB-DTPA-enhanced MRI and patient prognosis[J]. J Cancer Res Clin Oncol, 2019, 145(12): 2995-3003.
[26]
HuXJ, WangQ, HuangGB, et al. Gadoxetic acid-enhanced MRI-based radiomics signature: a potential imaging biomarker for identifying cytokeratin 19-positive hepatocellular carcinoma[J]. Comput Math Methods Med, 2023, 2023: 5424204.
[27]
LiuJQ, WangJ, HuangXL, et al. A radiomics model based on magnetic resonance imaging to predict cytokeratin 7/19 expression and liver fluke infection of hepatocellular carcinoma[J]. Sci Rep, 2023, 13(1): 17553.
[28]
ZhangLQ, ZhouHS, ZhangXQ, et al. A radiomics nomogram for predicting cytokeratin 19-positive hepatocellular carcinoma: a two-center study[J]. Front Oncol, 2023, 13: 1174069.
ShenL, LinZC, HuangQM. Relay backpropagation for effective learning of deep convolutional neural networks[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing, 2016: 467-82.
[31]
DutaIC, LiuL, ZhuF, et al. Improved residual networks for image and video recognition[C]//2020 25th International Conference on Pattern Recognition (ICPR). Milan, Italy. IEEE, 2021: 9415-22.
[32]
JeongHT, KimMJ, KimYE, et al. MRI features of hepatocellular carcinoma expressing progenitor cell markers[J]. Liver Int, 2012, 32(3): 430-40.
[33]
AnC, KimDW, ParkYN, et al. Single hepatocellular carcinoma: preoperative MR imaging to predict early recurrence after curative resection[J]. Radiology, 2015, 276(2): 433-43.
[34]
MaWJ, WangHY, TengLS. Correlation analysis of preoperative serum alpha-fetoprotein (AFP) level and prognosis of hepatocellular carcinoma (HCC) after hepatectomy[J]. World J Surg Oncol, 2013, 11: 212.