Deep learning-based arterial subtraction images improve the detection of LR-TR algorithm for viable HCC on extracellular agents-enhanced MRI.

PURPOSE: To determine the role of deep learning-based arterial subtraction images in viability assessment on extracellular agents-enhanced MRI using LR-TR algorithm. METHODS: Patients diagnosed with HCC who underwent locoregional therapy were retrospectively collected. We constructed a deep learning-based subtraction model and automatically generated arterial subtraction images. Two radiologists evaluated LR-TR category on ordinary images and then evaluated again on ordinary images plus arterial subtraction images after a 2-month washout period. The reference standard for viability was tumor stain on the digital subtraction hepatic angiography within 1 month after MRI. RESULTS: 286 observations of 105 patients were ultimately enrolled. 157 observations were viable and 129 observations were nonviable according to the reference standard. The sensitivity and accuracy of LR-TR algorithm for detecting viable HCC significantly increased with the application of arterial subtraction images (87.9% vs. 67.5%, p < 0.001; 86.4% vs. 75.9%, p < 0.001). And the specificity slightly decreased without significant difference when the arterial subtraction images were added (84.5% vs. 86.0%, p = 0.687). The AUC of LR-TR algorithm significantly increased with the addition of arterial subtraction images (0.862 vs. 0.768, p < 0.001). The arterial subtraction images also improved inter-reader agreement (0.857 vs. 0.727). CONCLUSION: Extended application of deep learning-based arterial subtraction images on extracellular agents-enhanced MRI can increase the sensitivity of LR-TR algorithm for detecting viable HCC without significant change in specificity.

The timing phase affected the inconsistency of APHE subtypes of liver observations in patients at risk for HCC on the multi-hepatic arterial phase imaging.

OBJECTIVE: To investigate whether liver observations in patients at risk for hepatocellular carcinoma (HCC) display inconsistent arterial phase hyperenhancement (APHE) subtypes on the multi-hepatic arterial phase imaging (mHAP) and to further investigate factors affecting inconsistent APHE subtype of observations on mHAP imaging. METHODS: From April 2018 to June 2021, a total of 141 patients at high risk of HCC with 238 liver observations who underwent mHAP MRI acquisitions were consecutively included in this retrospective study. Two experienced radiologists reviewed individual arterial phase imaging independently and assessed the enhancement pattern of each liver observation according to LI-RADS. Another two experienced radiologists identified and recorded the genuine timing phase of each phase independently. When a disagreement appeared between the two radiologists, another expert participated in the discussion to get a final decision. A separate descriptive analysis was used for all observations scored APHE by the radiologists. The Kappa coefficient was used to determine the agreement between the two radiologists. Univariate analysis was performed to investigate the factors affecting inconsistent APHE subtype of liver observations on mHAP imaging. RESULTS: The interobserver agreement was substantial to almost perfect agreement on the assessment of timing phase (κ = 0.712-0.887) and evaluation of APHE subtype (κ = 0.795-0.901). A total of 87.8% (209/238) of the observations showed consistent nonrim APHE and 10.2% (24/238) of the observations showed consistent rim APHE on mHAP imaging. A total of 2.1% (5/238) of the liver observations were considered inconsistent APHE subtypes, and all progressed nonrim to rim on mHAP imaging. 87.9% (124/141) of the mHAP acquisitions were all arterial phases and 12.1% (17/141) of the mHAP acquisitions obtained both the arterial phase and portal venous phase. Univariate analysis was performed and found that the timing phase of mHAP imaging affected the consistency of APHE subtype of liver observations. When considering the timing phase and excluding the portal venous phase acquired by mHAP imaging, none of the liver observations showed inconsistent APHE subtypes on mHAP imaging. CONCLUSION: The timing phase which mHAP acquisition contained portal venous phase affected the inconsistency of APHE subtype of liver observations on mHAP imaging. When evaluating the APHE subtype of liver observations, it's necessary to assess the timing of each phase acquired by the mHAP technique at first.

Preoperative evaluation of microvascular invasion in hepatocellular carcinoma with a radiological feature-based nomogram: a bi-centre study.

PURPOSE: To develop a nomogram for preoperative assessment of microvascular invasion (MVI) in hepatocellular carcinoma (HCC) based on the radiological features of enhanced CT and to verify two imaging techniques (CT and MRI) in an external centre. METHOD: A total of 346 patients were retrospectively included (training, n = 185, CT images; external testing 1, n = 90, CT images; external testing 2, n = 71, MRI images), including 229 MVI-negative patients and 117 MVI-positive patients. The radiological features and clinical information of enhanced CT images were analysed, and the independent variables associated with MVI in HCC were determined by logistic regression analysis. Then, a nomogram prediction model was constructed. External validation was performed on CT (n = 90) and MRI (n = 71) images from another centre. RESULTS: Among the 23 radiological and clinical features, size, arterial peritumoral enhancement (APE), tumour margin and alpha-fetoprotein (AFP) were independent influencing factors for MVI in HCC. The nomogram integrating these risk factors had a good predictive effect, with AUC, specificity and sensitivity values of 0.834 (95% CI: 0.774-0.895), 75.0% and 83.5%, respectively. The AUC values of external verification based on CT and MRI image data were 0.794 (95% CI: 0.700-0.888) and 0.883 (95% CI: 0.807-0.959), respectively. No statistical difference in AUC values among training set and testing sets was found. CONCLUSION: The proposed nomogram prediction model for MVI in HCC has high accuracy, can be used with different imaging techniques, and has good clinical applicability.

[Screen of key characteristic genes of nasopharyngeal carcinoma (NPC) base on machine learning and analysis of their correlation with immune cells].

Objective Machine learning was used to screen the key characteristic genes of nasopharyngeal carcinoma (NPC) and analyze their correlation with immune cells. Methods Download the NPC training datasets (GSE12452 and GSE13597) and the validation dataset (GSE53819) from the Gene Expression Omnibus (GEO). Firstly, the training data sets were merged and screened for differentially expressed genes (DEGs); Secondly, the DEGs were analyzed by gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), gene set enrichment analysis (GSEA), and immune cell infiltration analysis. Next, the least absolute shrinkage and selection operator (LASSO) and support vector machine (SVM) algorithms were used to identify NPC-related genes in the training datasets and examined in the validation dataset, to further identify key genes using the area under curve (AUC) of receiver operating characteristic curve (ROC); Finally, the correlation between the key genes and immune cells was analyzed. Results A total of 55 DEGs were obtained, including 43 down-regulated genes and 12 up-regulated genes. The GO functions were enriched in humoral immune response, cell differentiation, neutrophil activation and chemokine receptor binding. The KEGG were mainly enriched in the IL-17 signaling pathway. The GSEA was enriched in cell cycle, extracellular matrix receptor interactions, cancer pathways and DNA replication. Immune infiltration analysis showed that the expression of naive B cells, memory B cells, and resting memory CD4+ T cells was significantly lower in NPC, while CD8+ T cells, naive CD4+ T cells, activated memory CD4+ T cells, follicular helper T cells, M0 macrophages and M1 macrophages were highly expressed in NPC. Among the feature genes screened by LASSO and SVM, only CCDC19, LAMB1, SPAG6 and RAD51AP1 genes' AUC were greater than 0.9 in both the training and validation datasets and were closely associated with immune cell infiltration. Conclusion The key genes CCDC19, LAMB1, SPAG6 and RAD51AP1 in NPC development are screened by machine learning algorithms, and are closely associated with immune cell infiltration.