Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma

Abstract Objectives The potential of medical imaging to non-invasively assess intratumoral heterogeneity (ITH) is increasingly being recognized. This study aimed to investigate the value of the ITH-based deep learning model for preoperative prediction of histopathologic grade in hepatocellular carci...

Full description

Saved in:
Bibliographic Details
Main Authors: Shaoming Song, Gong Zhang, Zhiyuan Yao, Ruiqiu Chen, Kai Liu, Tianchen Zhang, Guineng Zeng, Zizheng Wang, Rong Liu
Format: Article
Language:English
Published: BMC 2025-03-01
Series:BMC Cancer
Subjects:
Online Access:https://doi.org/10.1186/s12885-025-13781-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850094989714915328
author Shaoming Song
Gong Zhang
Zhiyuan Yao
Ruiqiu Chen
Kai Liu
Tianchen Zhang
Guineng Zeng
Zizheng Wang
Rong Liu
author_facet Shaoming Song
Gong Zhang
Zhiyuan Yao
Ruiqiu Chen
Kai Liu
Tianchen Zhang
Guineng Zeng
Zizheng Wang
Rong Liu
author_sort Shaoming Song
collection DOAJ
description Abstract Objectives The potential of medical imaging to non-invasively assess intratumoral heterogeneity (ITH) is increasingly being recognized. This study aimed to investigate the value of the ITH-based deep learning model for preoperative prediction of histopathologic grade in hepatocellular carcinoma (HCC). Materials and methods A total of 858 patients from primary cohort and two external cohorts were included. 3.0T or 1.5T axial portal venous phase MRI images were collected. We conducted radiomics feature-driven K-means clustering for automatic partition to reveal ITH. 2.5D and 3D deep learning models based on ResNet architecture were trained to extract deep learning hidden features of each subregion. The selected features were used to train Random Forest classifier, which constructed the feature-fusion model. Results The extracted voxel-level radiomics features were unsupervised clustered by K-means to generate three subregions. In the 2.5D deep learning, the feature-fusion model based on ITH had superior predictive efficacy than the whole-tumor model (AUC: 0.82 vs. 0.72; p = 0.004). Even in the validation and external test sets, this model maintained a high AUC of 0.78–0.83, and net reclassification indices indicated that it could improve prediction by 25–28%. Regarding the prognostic value, overall survival (OS) and recurrence-free survival (RFS) could be significantly stratified by the 2.5D feature-fusion model, and multivariable Cox regressions indicated its signature was identified as a risk predictor for OS and RFS (p < 0.05). Conclusion The ITH-based feature-fusion model provided a non-invasive method for classifying tumor differentiation in HCC, which may serve as a promising strategy for stratification management.
format Article
id doaj-art-198d693df4c8434b9546cbfe75bd1a1a
institution DOAJ
issn 1471-2407
language English
publishDate 2025-03-01
publisher BMC
record_format Article
series BMC Cancer
spelling doaj-art-198d693df4c8434b9546cbfe75bd1a1a2025-08-20T02:41:32ZengBMCBMC Cancer1471-24072025-03-0125111310.1186/s12885-025-13781-1Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinomaShaoming Song0Gong Zhang1Zhiyuan Yao2Ruiqiu Chen3Kai Liu4Tianchen Zhang5Guineng Zeng6Zizheng Wang7Rong Liu8The First School of Clinical Medicine, Lanzhou UniversityFaculty of Hepatopancreatobiliary Surgery, The First Medical Center of Chinese PLA General HospitalFaculty of Hepatopancreatobiliary Surgery, The First Medical Center of Chinese PLA General HospitalThe First School of Clinical Medicine, Lanzhou UniversityFaculty of Hepatopancreatobiliary Surgery, The First Medical Center of Chinese PLA General HospitalThe First School of Clinical Medicine, Lanzhou UniversityFaculty of Hepatopancreatobiliary Surgery, The First Medical Center of Chinese PLA General HospitalDepartment of Hepatobiliary Surgery, Senior Department of Hepatology, The Fifth Medical Center of Chinese PLA General HospitalThe First School of Clinical Medicine, Lanzhou UniversityAbstract Objectives The potential of medical imaging to non-invasively assess intratumoral heterogeneity (ITH) is increasingly being recognized. This study aimed to investigate the value of the ITH-based deep learning model for preoperative prediction of histopathologic grade in hepatocellular carcinoma (HCC). Materials and methods A total of 858 patients from primary cohort and two external cohorts were included. 3.0T or 1.5T axial portal venous phase MRI images were collected. We conducted radiomics feature-driven K-means clustering for automatic partition to reveal ITH. 2.5D and 3D deep learning models based on ResNet architecture were trained to extract deep learning hidden features of each subregion. The selected features were used to train Random Forest classifier, which constructed the feature-fusion model. Results The extracted voxel-level radiomics features were unsupervised clustered by K-means to generate three subregions. In the 2.5D deep learning, the feature-fusion model based on ITH had superior predictive efficacy than the whole-tumor model (AUC: 0.82 vs. 0.72; p = 0.004). Even in the validation and external test sets, this model maintained a high AUC of 0.78–0.83, and net reclassification indices indicated that it could improve prediction by 25–28%. Regarding the prognostic value, overall survival (OS) and recurrence-free survival (RFS) could be significantly stratified by the 2.5D feature-fusion model, and multivariable Cox regressions indicated its signature was identified as a risk predictor for OS and RFS (p < 0.05). Conclusion The ITH-based feature-fusion model provided a non-invasive method for classifying tumor differentiation in HCC, which may serve as a promising strategy for stratification management.https://doi.org/10.1186/s12885-025-13781-1Intratumoral heterogeneityDeep learningHistopathologic gradeMagnetic resonance imagingHepatocellular carcinoma
spellingShingle Shaoming Song
Gong Zhang
Zhiyuan Yao
Ruiqiu Chen
Kai Liu
Tianchen Zhang
Guineng Zeng
Zizheng Wang
Rong Liu
Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma
BMC Cancer
Intratumoral heterogeneity
Deep learning
Histopathologic grade
Magnetic resonance imaging
Hepatocellular carcinoma
title Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma
title_full Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma
title_fullStr Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma
title_full_unstemmed Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma
title_short Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma
title_sort deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma
topic Intratumoral heterogeneity
Deep learning
Histopathologic grade
Magnetic resonance imaging
Hepatocellular carcinoma
url https://doi.org/10.1186/s12885-025-13781-1
work_keys_str_mv AT shaomingsong deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT gongzhang deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT zhiyuanyao deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT ruiqiuchen deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT kailiu deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT tianchenzhang deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT guinengzeng deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT zizhengwang deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma
AT rongliu deeplearningbasedonintratumoralheterogeneitypredictshistopathologicgradeofhepatocellularcarcinoma