Preoperative Prediction of Microvascular Invasion in Hepatocellular Carcinoma: Initial Application of a Radiomic Algorithm Based on Grayscale Ultrasound Images

Abstract

Objectives: To establish a radiomic algorithm based on grayscale ultrasound images and to make preoperative predictions of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) patients. Methods: In this retrospective study, 322 cases of histopathologically confirmed HCC lesions were included. The classifications based on preoperative grayscale ultrasound images were performed in two stages: (1) classifier #1, MVI-negative and MVI-positive cases; (2) classifier #2, MVI-positive cases were further classified as M1 or M2 cases. The gross-tumoral region (GTR) and peri-tumoral region (PTR) signatures were combined to generate gross-and peri-tumoral region (GPTR) radiomic signatures. The optimal radiomic signatures were further incorporated with vital clinical information. Multivariable logistic regression was used to build radiomic models. Results: Finally, 1,595 radiomic features were extracted from each HCC lesion. At the classifier #1 stage, the radiomic signatures based on features of GTR, PTR, and GPTR showed area under the curve (AUC) values of 0.708 (95% CI, 0.603-0.812), 0.710 (95% CI, 0.609-0.811), and 0.726 (95% CI, 0.625-0.827), respectively. Upon incorporation of vital clinical information, the AUC of the GPTR radiomic algorithm was 0.744 (95% CI, 0.646-0.841). At the classifier #2 stage, the AUC of the GTR radiomic signature was 0.806 (95% CI, 0.667-0.944). Conclusions: Our radiomic algorithm based on grayscale ultrasound images has potential value to facilitate preoperative prediction of MVI in HCC patients. The GTR radiomic signature may be helpful for further discriminating between M1 and M2 levels among MVI-positive patients.

Prediction of Microvascular Invasion of Hepatocellular Carcinoma Based on Contrast-Enhanced MR and 3D Convolutional Neural Networks

“…Xu et al ( 15 ) recently assessed the radiomics characteristics for MVI prediction, and demonstrated that the radiomics signatures of contrast-enhanced CT were less important than the radiological features, with the AUC value from 0.787 to 0.841. In addition, the radiomics nomogram has also been used in the ultrasound images for MVI prediction with the reported AUC value of 0.731 ( 16 ) and 0.806 ( 17 ), respectively. Comparatively, our proposed deep learning framework with 3D CNN yielded the performance with the AUC value of 0.926, which is better than the reported radiomics approaches.…”

“…Xu et al ( 15 ) recently analyzed the contrast-enhanced CT based on radiomics analysis to predict MVI and outcome in HCC, and demonstrated that the combination of radiological and imaging radiomics features could produce better performance in predicting MVI. In addition, there also have been some reports about radiomics predicting MVI based on ultrasound images ( 16 , 17 ). As pointed out by the recent study ( 8 ), researchers currently construct radiomics models based on single modality image data, and the use of multimodality for MVI prediction has not to be investigated.…”

AI-Based Radiological Imaging for HCC: Current Status and Future of Ultrasound

“…For the former, although the sample size was small (47 patients with HCC), a radiomics-based study including B-mode, share wave elastography (SWE) and viscosity imaging demonstrated high diagnostic ability (AUC 0.94) to predict Ki-67, which is a marker to indicate the poor prognosis of several malignant diseases [ 48 ]. For the latter, two retrospective studies using B-mode US findings showed AUCs of 0.731 and 0.726 (0.806 for differentiation between M1 and M2) in predicting MVI [ 49 , 80 ]. In addition, Yao et al reported a higher diagnostic ability (AUC 0.98) achieved by radiomics-based multimodal US images [ 48 ].…”

Application of artificial intelligence in the diagnosis and treatment of hepatocellular carcinoma: A review

“…However, these techniques involve radiological exposure, and are laborious to perform and costly. Recently, Dong et al[ 30 ] published a study using radiomic algorithms based on grayscale ultrasound images to elaborate radiomic signatures with the potential to aid in the prediction of VMI, with promising results. Ji et al[ 31 ] created predictive models for recurrence after surgical resection using radiomic techniques to analyze contrast-enhanced CT images, with a C-index of 0.633-0.699.…”