CEUS-Based Radiomics Can Show Changes in Protein Levels in Liver Metastases After Incomplete Thermal Ablation.

Haiwei Bao,Junyan Zhu,Haiyang Xie,Fen Chen,Ting Chen

doi:10.3389/fonc.2021.694102

Abstract

ObjectiveTo investigate the ability of contrast-enhanced ultrasound (CEUS)-based radiomics combined with machine learning to detect early protein changes after incomplete thermal ablation.MethodsHCT-26 colorectal adenoma cells were engrafted into the livers of 80 mice, which were randomly divided into 4 groups for palliative laser ablation. Changes in heat shock protein (HSP) and apoptosis-related protein expression in the tumors were assessed. SCID mice subjected to CEUS and ultrasonography were divided into training (n=56) and test (n=24) datasets. Then, 102 features from seven feature groups were extracted. We use the least absolute shrinkage and selection operator (LASSO) feature selection method to fit the machine learning classifiers. The feature selection methods and four classifiers were combined to determine the best prediction model.ResultsThe areas under the receiver-operating characteristic curves (AUCs) of the classifiers in the test dataset ranged from 0.450 to 0.932 (median: 0.721). The best score was obtained from the model in which the omics data of CEUS was analyzed in the arterial phase by random forest (RF) classification.ConclusionsA machine learning model, in which radiomics characteristics are extracted by multimodal ultrasonography, can accurately, rapidly and noninvasively identify protein changes after ablation.

Highlights

Whether liver metastases can be inactivated in patients with liver metastases from colorectal cancer (CRLM) is a key issue influencing the survival and long-term tumor-free survival of patients [1,2,3]
The western blotting results confirmed that the HSP70 and HSP90a expression levels were significantly higher in experimental groups 3 and 4 than in control groups 1 and 2; the relative HSP70 and HSP90a expression levels in group 4 were higher than those in group 3 (p < 0.01) (Figures 4A, B)
The results show that the binary classification machine learning model trained by quantitative radiomics ultrasound data can distinguish changes in protein levels after incomplete ablation

Summary

Introduction

Whether liver metastases can be inactivated in patients with liver metastases from colorectal cancer (CRLM) is a key issue influencing the survival and long-term tumor-free survival of patients [1,2,3]. There is an urgent need for noninvasive imaging methods to reveal the molecular changes in the tumor before and after ablation [9]. In theory, imaging has the potential to reveal the tumor phenotype and molecular changes from the level of macroscopic characteristics of organ tissues to the level of local cell and molecular characteristics [12]. There are relatively few reports on multimodal ultrasound, including contrastenhanced ultrasound (CEUS), and there is no standard for omics exploration methods [21,22,23]. The main difficulty is the lack of universal ultrasound omics analysis methods, especially for CEUS data [24, 25]

Methods

Results

Discussion

Conclusion