Dataset Information

Preoperative identification of microvascular invasion in hepatocellular carcinoma by XGBoost and deep learning.

ABSTRACT:

Purpose

Microvascular invasion (MVI) is a valuable predictor of survival in hepatocellular carcinoma (HCC) patients. This study developed predictive models using eXtreme Gradient Boosting (XGBoost) and deep learning based on CT images to predict MVI preoperatively.

Methods

In total, 405 patients were included. A total of 7302 radiomic features and 17 radiological features were extracted by a radiomics feature extraction package and radiologists, respectively. We developed a XGBoost model based on radiomics features, radiological features and clinical variables and a three-dimensional convolutional neural network (3D-CNN) to predict MVI status. Next, we compared the efficacy of the two models.

Results

Of the 405 patients, 220 (54.3%) were MVI positive, and 185 (45.7%) were MVI negative. The areas under the receiver operating characteristic curves (AUROCs) of the Radiomics-Radiological-Clinical (RRC) Model and 3D-CNN Model in the training set were 0.952 (95% confidence interval (CI) 0.923-0.973) and 0.980 (95% CI 0.959-0.993), respectively (p?=?0.14). The AUROCs of the RRC Model and 3D-CNN Model in the validation set were 0.887 (95% CI 0.797-0.947) and 0.906 (95% CI 0.821-0.960), respectively (p?=?0.83). Based on the MVI status predicted by the RRC and 3D-CNN Models, the mean recurrence-free survival (RFS) was significantly better in the predicted MVI-negative group than that in the predicted MVI-positive group (RRC Model: 69.95 vs. 24.80 months, p?ConclusionThe RRC Model and 3D-CNN models showed considerable efficacy in identifying MVI preoperatively. These machine learning models may facilitate decision-making in HCC treatment but requires further validation.

SUBMITTER: Jiang YQ

PROVIDER: S-EPMC7873117 | biostudies-literature | 2021 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Preoperative identification of microvascular invasion in hepatocellular carcinoma by XGBoost and deep learning.

Jiang Yi-Quan YQ Cao Su-E SE Cao Shilei S Chen Jian-Ning JN Wang Guo-Ying GY Shi Wen-Qi WQ Deng Yi-Nan YN Cheng Na N Ma Kai K Zeng Kai-Ning KN Yan Xi-Jing XJ Yang Hao-Zhen HZ Huan Wen-Jing WJ Tang Wei-Min WM Zheng Yefeng Y Shao Chun-Kui CK Wang Jin J Yang Yang Y Chen Gui-Hua GH

Journal of cancer research and clinical oncology 20200827 3

<h4>Purpose</h4>Microvascular invasion (MVI) is a valuable predictor of survival in hepatocellular carcinoma (HCC) patients. This study developed predictive models using eXtreme Gradient Boosting (XGBoost) and deep learning based on CT images to predict MVI preoperatively.<h4>Methods</h4>In total, 405 patients were included. A total of 7302 radiomic features and 17 radiological features were extracted by a radiomics feature extraction package and radiologists, respectively. We developed a XGBoos ...[more]

PMID: 32852634

Similar Datasets

Project description:BackgroundRadiomics has emerged as a new approach that can help identify imaging information associated with tumor pathophysiology. We developed and validated a radiomics nomogram for preoperative prediction of microvascular invasion (MVI) in hepatocellular carcinoma (HCC).MethodsTwo hundred and eight patients with pathologically confirmed HCC (training cohort: n = 146; validation cohort: n = 62) who underwent preoperative gadoxetic acid-enhanced magnetic resonance (MR) imaging were included. Least absolute shrinkage and selection operator logistic regression was applied to select features and construct signatures derived from MR images. Univariate and multivariate analyses were used to identify the significant clinicoradiological variables and radiomics signatures associated with MVI, which were then incorporated into the predictive nomogram. The performance of the radiomics nomogram was evaluated by its calibration, discrimination, and clinical utility.ResultsHigher α-fetoprotein level (p = 0.046), nonsmooth tumor margin (p = 0.003), arterial peritumoral enhancement (p < 0.001), and the radiomics signatures of hepatobiliary phase (HBP) T1-weighted images (p < 0.001) and HBP T1 maps (p < 0.001) were independent risk factors of MVI. The predictive model that incorporated the clinicoradiological factors and the radiomic features derived from HBP images outperformed the combination of clinicoradiological factors in the training cohort (area under the curves [AUCs] 0.943 vs. 0.850; p = 0.002), though the validation did not have a statistical significance (AUCs 0.861 vs. 0.759; p = 0.111). The nomogram based on the model exhibited C-index of 0.936 (95% CI 0.895-0.976) and 0.864 (95% CI 0.761-0.967) in the training and validation cohort, fitting well in calibration curves (p > 0.05). Decision curve analysis further confirmed the clinical usefulness of the nomogram.ConclusionsThe nomogram incorporating clinicoradiological risk factors and radiomic features derived from HBP images achieved satisfactory preoperative prediction of the individualized risk of MVI in patients with HCC.

Project description:BackgroundAs an indicator of tumor invasiveness, microvascular invasion (MVI) is a crucial risk factor for postoperative relapse, metastasis, and unfavorable prognosis in hepatocellular carcinoma (HCC). Nevertheless, the genetic mechanisms underlying MVI, particularly for Chinese patients, remain mostly uncharted.MethodsWe applied deep targeted sequencing on 66 Chinese HCC samples. Focusing on the telomerase reverse transcriptase (TERT) promoter (TERTp) and TP53 co-mutation (TERTp+/TP53+) group, gene set enrichment analysis (GSEA) was used to explore the potential molecular mechanisms of the TERTp+/TP53+ group on tumor progression and metastasis. Additionally, we evaluated the tumor immune microenvironment of the TERTp+/TP53+ group in HCC using multiplex immunofluorescence (mIF) staining.ResultsAmong the 66 HCC samples, the mutated genes that mostly appeared were TERT, TP53, and CTNNB1. Of note, we found 10 cases with TERTp+/TP53+, of which nine were MVI-positive and one was MVI-negative, and there was a co-occurrence of TERTp and TP53 (p < 0.05). Survival analysis demonstrated that patients with the TERTp+/TP53+ group had lower the disease-free survival (DFS) (p = 0.028). GSEA results indicated that telomere organization, telomere maintenance, DNA replication, positive regulation of cell cycle, and negative regulation of immune response were significantly enriched in the TERTp+/TP53+ group (all adjusted p-values (p.adj) < 0.05). mIF revealed that the TERTp+/TP53+ group decreased CD8+ T cells infiltration (p = 0.25) and enhanced PDL1 expression (p = 0.55).ConclusionsTERTp+/TP53+ was significantly enriched in MVI-positive patients, leading to poor prognosis for HCC patients by promoting proliferation of HCC cell and inhibiting infiltration of immune cell surrounding HCC. TERTp+/TP53+ can be utilized as a potential indicator for predicting MVI-positive patients and poor prognosis, laying a preliminary foundation for further exploration of co-mutation in HCC with MVI and clinical treatment.

Project description:BackgroundMicrovascular invasion (MVI) is an independent risk factor for postoperative recurrence of hepatocellular carcinoma (HCC). To perform a meta-analysis to investigate the diagnostic performance of radiomics for the preoperative evaluation of MVI in HCC and the effect of potential factors.Materials and methodsA systematic literature search was performed in PubMed, Embase, and the Cochrane Library for studies focusing on the preoperative evaluation of MVI in HCC with radiomics methods. Data extraction and quality assessment of the retrieved studies were performed. Statistical analysis included data pooling, heterogeneity testing and forest plot construction. Meta-regression and subgroup analyses were performed to reveal the effect of potential explanatory factors [design, combination of clinical factors, imaging modality, number of participants, and Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) applicability risk] on the diagnostic performance.ResultsTwenty-two studies with 4,129 patients focusing on radiomics for the preoperative prediction of MVI in HCC were included. The pooled sensitivity, specificity and area under the receiver operating characteristic curve (AUC) were 84% (95% CI: 81, 87), 83% (95% CI: 78, 87) and 0.90 (95% CI: 0.87, 0.92). Substantial heterogeneity was observed among the studies (I²=94%, 95% CI: 88, 99). Meta-regression showed that all investigative covariates contributed to the heterogeneity in the sensitivity analysis (P < 0.05). Combined clinical factors, MRI, CT and number of participants contributed to the heterogeneity in the specificity analysis (P < 0.05). Subgroup analysis showed that the pooled sensitivity, specificity and AUC estimates were similar among studies with CT or MRI.ConclusionRadiomics is a promising noninvasive method that has high preoperative diagnostic performance for MVI status. Radiomics based on CT and MRI had a comparable predictive performance for MVI in HCC. Prospective, large-scale and multicenter studies with radiomics methods will improve the diagnostic power for MVI in the future.Systematic review registrationhttps://www.crd.york.ac.uk/prospero/display_record.php?RecordID=259363, identifier CRD42021259363.

Project description:PurposesThis study aimed to establish a predictive model of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) by contrast-enhanced computed tomography (CT), which relied on a combination of machine learning approach and imaging features covering Liver Imaging and Reporting and Data System (LI-RADS) features.MethodsThe retrospective study included 279 patients with surgery who underwent preoperative enhanced CT. They were randomly allocated to training set, validation set, and test set (167 patients vs. 56 patients vs. 56 patients, respectively). Significant imaging findings for predicting MVI were identified through the Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression method. Predictive models were performed by machine learning algorithm, support vector machine (SVM), in the training set and validation set, and evaluated in the test set. Further, a combined model adding clinical findings to the radiologic model was developed. Based on the LI-RADS category, subgroup analyses were conducted.ResultsWe included 116 patients with MVI which were diagnosed through pathological confirmation. Six imaging features were selected about MVI prediction: four LI-RADS features (corona enhancement, enhancing capsule, non-rim aterial phase hyperehancement, tumor size) and two non-LI-RADS features (internal arteries, non-smooth tumor margin). The radiological feature with the best accuracy was corona enhancement followed by internal arteries and tumor size. The accuracies of the radiological model and combined model were 0.725-0.714 and 0.802-0.732 in the training set, validation set, and test set, respectively. In the LR-4/5 subgroup, a sensitivity of 100% and an NPV of 100% were obtained by the high-sensitivity threshold. A specificity of 100% and a PPV of 100% were acquired through the high specificity threshold in the LR-M subgroup.ConclusionA combination of LI-RADS features and non-LI-RADS features and serum alpha-fetoprotein value could be applied as a preoperative biomarker for predicting MVI by the machine learning approach. Furthermore, its good performance in the subgroup by LI-RADS category may help optimize the management of HCC patients.

Dataset Information

Preoperative identification of microvascular invasion in hepatocellular carcinoma by XGBoost and deep learning.

Purpose

Methods

Results

Publications

Preoperative identification of microvascular invasion in hepatocellular carcinoma by XGBoost and deep learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets