Prediction of microvascular invasion in hepatocellular carcinoma based on preoperative Gd-EOB-DTPA-enhanced MRI: Comparison of predictive performance among 2D, 2D-expansion and 3D deep learning models

Abstract

PurposeTo evaluate and compare the predictive performance of different deep learning models using gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)-enhanced MRI in predicting microvascular invasion (MVI) in hepatocellular carcinoma.MethodsThe data of 233 patients with pathologically confirmed hepatocellular carcinoma (HCC) treated at our hospital from June 2016 to June 2021 were retrospectively analyzed. Three deep learning models were constructed based on three different delineate methods of the region of interest (ROI) using the Darwin Scientific Research Platform (Beijing Yizhun Intelligent Technology Co., Ltd., China). Manual segmentation of ROI was performed on the T1-weighted axial Hepatobiliary phase images. According to the ratio of 7:3, the samples were divided into a training set (N=163) and a validation set (N=70). The receiver operating characteristic (ROC) curve was used to evaluate the predictive performance of three models, and their sensitivity, specificity and accuracy were assessed.ResultsAmong 233 HCC patients, 109 were pathologically MVI positive, including 91 men and 18 women, with an average age of 58.20 ± 10.17 years; 124 patients were MVI negative, including 93 men and 31 women, with an average age of 58.26 ± 10.20 years. Among three deep learning models, 2D-expansion-DL model and 3D-DL model showed relatively good performance, the AUC value were 0.70 (P=0.003) (95% CI 0.57–0.82) and 0.72 (P<0.001) (95% CI 0.60–0.84), respectively. In the 2D-expansion-DL model, the accuracy, sensitivity and specificity were 0.7143, 0.739 and 0.688. In the 3D-DL model, the accuracy, sensitivity and specificity were 0.6714, 0.800 and 0.575, respectively. Compared with the 3D-DL model (based on 3D-ResNet), the 2D-DL model is smaller in scale and runs faster. The frames per second (FPS) for the 2D-DL model is 244.7566, which is much larger than that of the 3D-DL model (73.3374).ConclusionThe deep learning model based on Gd-EOB-DTPA-enhanced MRI could preoperatively evaluate MVI in HCC. Considering that the predictive performance of 2D-expansion-DL model was almost the same as the 3D-DL model and the former was relatively easy to implement, we prefer the 2D-expansion-DL model in practical research.

Leveraging radiomics and AI for precision diagnosis and prognostication of liver malignancies

“…Several studies using AI or radiomic features extracted from gadoxetic acid-enhanced MRI, dynamic contrast enhanced MR, or contrast enhanced CT images tried to predict microvascular invasion in HCC and mass-forming CC ( 36 ). The AUCs ranged from 0.75 to 0.98 with most of the studies achieving AUCs higher than 0.85 ( 36 – 47 ). Notably, studies focused on peritumoral areas within the 5 cm to 10 cm range.…”

“…One study underscored that patients without MVI experienced significantly prolonged recurrence-free survival (RFS). Validation sets were incorporated in all studies ( 36 – 47 ). As mentioned above, accurately predicting MVI before surgery can significantly influence surgical planning, including decisions regarding the extent of resection or the suitability of ablation treatments.…”

MRI‐Based Radiomics and Deep Learning in Biological Characteristics and Prognosis of Hepatocellular Carcinoma: Opportunities and Challenges

“…66 Some studies compared the predictive performance of different deep learning models between inputs or backbone in predicting MVI. 67 Such comparisons can be tedious as the models are independent, requiring ablation experiments to help humans truly understand the possible mechanisms. A multi-center and prospective validation study revealed that an efficient deep learning model based on EOB-enhanced MRI achieved superior performance in predicting MVI.…”

“…Gao et al proposed an ensemble learning algorithm based on the different degree and attention mechanisms, combined with radiomics for MVI prediction with an AUC of 0.83 66 . Some studies compared the predictive performance of different deep learning models between inputs or backbone in predicting MVI 67 . Such comparisons can be tedious as the models are independent, requiring ablation experiments to help humans truly understand the possible mechanisms.…”

Advancing Mixed Reality Digital Twins Through 3D Reconstruction of Fresh Produce

“…In the field of cell biology, 3D models based on embryoid body differentiation show advantages over 2D monolayer models, fostering enhanced cell interactions and promoting the induction of hematopoietic lineages [10]. In medical imaging, 3D deep learning models have demonstrated enhanced predictive performance over 2D models, specifically in predicting microvascular invasion in hepatocellular carcinoma [11]. In precision agriculture, the use of 3D reconstruction, such as laser scanning technology for maize plants, facilitates non-destructive crop monitoring and phenotypic analysis, revolutionizing agricultural practices [12].…”