Deep learning-based and BI-RADS guided radiomics model for automatic tumor-infiltrating lymphocytes evaluation in breast cancer.

OBJECTIVES: To investigate an interpretable radiomics model consistent with clinical decision-making process and realize automatic prediction of tumor-infiltrating lymphocytes (TILs) levels in breast cancer (BC) from ultrasound (US) images. METHODS: A total of 378 patients with invasive BC confirmed by pathological results were retrospectively enrolled in this study. Radiomics features were extracted guided by the BI-RADS lexicon from the regions of interest(ROIs) segmented with deep learning models. After features selected using the least absolute shrinkage and selection operator(LASSO) regression, four machine learning classifiers were used to establish the radiomics signature(Rad-score). Then, the integrated model was developed on the basis of the best Rad-score incorporating the independent clinical factors for TILs levels prediction. RESULTS: Tumors were segmented using the deep learning models with accuracy of 97.2%, sensitivity of 93.4%, specificity of 98.1%, and the posterior areas were also obtained. Eighteen morphology and texture related features were extracted from the ROIs and fourteen features were selected to construct the Rad-score models. Combined with independent clinical characteristics, the integrated model achieved an area under the curve (AUC) of 0.889(95% CI,0.739,0.990) in the validation cohort and outperformed the traditional radiomics model with AUC of 0.756(0.649-0862) depended on hundreds of feature items. CONCLUSIONS: This study established a promising model for TILs levels prediction with numerable interpretable features and showed great potential to help decision-making and clinical applications. ADVANCES IN KNOWLEDGE: Imaging-based biomarkers has provides non-invasive ways for TILs levels evaluation in BC. Our model combining the BI-RADS guided radiomics features and clinical data outperformed the traditional radiomics approaches.

MFMSNet: A Multi-frequency and Multi-scale Interactive CNN-Transformer Hybrid Network for breast ultrasound image segmentation.

Breast tumor segmentation in ultrasound images is fundamental for quantitative analysis and plays a crucial role in the diagnosis and treatment of breast cancer. Recently, existing methods have mainly focused on spatial domain implementations, with less attention to the frequency domain. In this paper, we propose a Multi-frequency and Multi-scale Interactive CNN-Transformer Hybrid Network (MFMSNet). Specifically, we utilize Octave convolutions instead of conventional convolutions to effectively separate high-frequency and low-frequency components while reducing computational complexity. Introducing the Multi-frequency Transformer block (MF-Trans) enables efficient interaction between high-frequency and low-frequency information, thereby capturing long-range dependencies. Additionally, we incorporate Multi-scale interactive fusion module (MSIF) to merge high-frequency feature maps of different sizes, enhancing the emphasis on tumor edges by integrating local contextual information. Experimental results demonstrate the superiority of our MFMSNet over seven state-of-the-art methods on two publicly available breast ultrasound datasets and one thyroid ultrasound dataset. In the evaluation of MFMSNet, tests were conducted on the BUSI, BUI, and DDTI datasets, comprising 130 images (BUSI), 47 images (BUI), and 128 images (DDTI) in the respective test sets. Employing a five-fold cross-validation approach, the obtained dice coefficients are as follows: 83.42 % (BUSI), 90.79 % (BUI), and 79.96 % (DDTI). The code is available at https://github.com/wrc990616/MFMSNet.

The value of Demetics ultrasound-assisted diagnosis system in diagnosis of breast lesions and in assessment Ki-67 status of breast cancer.

PURPOSE: This study aims to explore the diagnostic efficiency of the Demetics for breast lesions and assessment of Ki-67 status. MATERIAL: This retrospective study included 291 patients. Three combined methods (method 1: upgraded BI-RADS when Demetics classified the breast lesion as malignant; method 2: downgraded BI-RADS when Demetics classified the breast lesion as benign; method 3: BI-RADS was upgraded or downgraded according to Demetrics' diagnosis) were used to compare the diagnostic efficiency of two radiologists with different seniority before and after using Demetics. The correlation between the visual heatmap by Demetics and the Ki-67 expression level of breast cancer was explored. RESULTS: The sensitivity, specificity, and area under curve (AUC) of diagnosis by Demetics, junior radiologist and senior radiologist were 89.5%, 83.1%, 0.863; 76.9%, 82.4%, 0.797 and 81.1%, 89.9%, 0.855, respectively. Method 1 was the best for senior radiologist, which increased AUC from 0.855 to 0.884. For junior radiologist, Method 3 was the best method, improving sensitivity (88.8% vs. 76.9%) and specificity (87.2% vs. 82.4%). Demetics paid more attention to the peripheral area of breast cancer with high expression of Ki-67. CONCLUSION: Demetics has shown good diagnostic efficiency in the assisted diagnosis of breast lesions and is expected to further distinguish Ki-67 status of breast cancer.

Evaluation of Breast Cancer Tumor-Infiltrating Lymphocytes on Ultrasound Images Based on a Novel Multi-Cascade Residual U-Shaped Network.

OBJECTIVE: Breast cancer has become the leading cancer of the 21st century. Tumor-infiltrating lymphocytes (TILs) have emerged as effective biomarkers for predicting treatment response and prognosis in breast cancer. The work described here was aimed at designing a novel deep learning network to assess the levels of TILs in breast ultrasound images. METHODS: We propose the Multi-Cascade Residual U-Shaped Network (MCRUNet), which incorporates a gray feature enhancement (GFE) module for image reconstruction and normalization to achieve data synergy. Additionally, multiple residual U-shaped (RSU) modules are cascaded as the backbone network to maximize the fusion of global and local features, with a focus on the tumor's location and surrounding regions. The development of MCRUNet is based on data from two hospitals and uses a publicly available ultrasound data set for transfer learning. RESULTS: MCRUNet exhibits excellent performance in assessing TILs levels, achieving an area under the receiver operating characteristic curve of 0.8931, an accuracy of 85.71%, a sensitivity of 83.33%, a specificity of 88.64% and an F1 score of 86.54% in the test group. It outperforms six state-of-the-art networks in terms of performance. CONCLUSION: The MCRUNet network based on breast ultrasound images of breast cancer patients holds promise for non-invasively predicting TILs levels and aiding personalized treatment decisions.

Methylated SEPT9 combined with AFP and PIVKA-II is effective for the detection of HCC in high-risk population.

BACKGROUND: The methylation SEPT9 (mSEPT9) appeared to be effective for hepatocellular carcinoma (HCC) detection. However, its performance in high-risk population has not been validated. We designed a pilot study and aimed to investigate the performance of mSEPT9, AFP, PIVKA-II and their combination in hepatic cirrhosis (HC) population. METHODS: A training cohort was established including 103 HCC and 114 HC patients. 10 ml blood was collected from each patient with K2EDTA tubes, and 3-4 ml plasma was extracted for subsequent tests. The performance of mSEPT9, AFP, PIVKA-II and their combination was optimized by the training cohort. Test performance was prospectively validated with a validation cohort, including 51 HCC and 121 HC patients. RESULTS: At the optimal thresholds in the training cohort, the sensitivity, specificity and area under curve (AUC) was 72.82%, 89.47%, 0.84, and 48.57%, 89.92%, 0.79, and 63.64%, 95.95%, 0.79 for mSEPT9, AFP and PIVKA-II, respectively. The combined test significantly increased the sensitivity to 84.47% (P < 0.05) at the specificity of 86.84% with an AUC of 0.91. Stage-dependent performance was observed with all single markers and their combination in plasma marker levels, positive detection rate (PDR) and AUC. Moderate correlation was found between mSEPT9 and AFP plasma levels (r = 0.527, P < 0.0001). Good complementarity was found between any two of the three markers, providing optimal sensitivity in HCC detection when used in combination. Subsequent validation achieved a sensitivity, specificity and AUC of 65.31%, 92.86%, 0.80, and 44.24%, 89.26%, 0.75, and 62.22%, 95.27%, 0.78 for mSEPT9, AFP and PIVKA-II, respectively. The combined test yielded a significantly increased sensitivity of 84.00% (P < 0.05) at 85.57% specificity, with an AUC at 0.89. CONCLUSIONS: The performance was optimal by the combination of mSEPT9, AFP, PIVKA-II compared with any single marker, and the combination may be effective for HCC opportunistic screening in HC population.

Analysis of angiogenesis-related subtypes of hepatocellular carcinoma and tumor microenvironment infiltration feature in hepatocellular carcinoma

Purpose Hepatocellular carcinoma (HCC) is a highly vascularized tumor, and angiogenesis plays an important role in its progression. However, the role of angiogenesis in cell infiltration in the tumor microenvironment (TME) remains unclear. Methods We evaluated the associations of 35 angiogenesis-related genes (ARGs) with the clinicopathological features of 816 HCC patients. In addition, we assessed the associations between the ARGs and TME cell infiltration. A nomogram was constructed to determine the prognostic value of ARGs for HCC. The ARG score was used to distinguish angiogenic subtypes of HCC, and its usefulness for predicting the prognosis and treatment response of HCC patients was evaluated. Results We distinguished three ARG clusters differing in terms of TME cell infiltration, immune cell activation status, clinicopathological features, and clinical outcomes. There were significant associations of ARG expression with tumor immunity, the epithelial–mesenchymal transition (EMT), and transforming growth factor-β expression. An ARG score model was constructed to generate a risk score for each patient based on differentially expressed genes between clusters. Furthermore, a high ARG score was associated with high expression of CTLA-4 and PD-L1/PD-1, and a low Tumor Immune Dysfunction and Exclusion score, indicating the usefulness of the ARG score for selecting patients for immunotherapy. Considering the relationship between ARGs and tumor immunity, immunotherapy combined with vascular-targeted therapy may be the best treatment for HCC. Conclusions ARGs play an important role in TME diversity and complexity in HCC patients. The ARG score of HCC predicts TME invasion and can guide immunotherapy (AU)

Deep Learning with Transformer or Convolutional Neural Network in the Assessment of Tumor-Infiltrating Lymphocytes (TILs) in Breast Cancer Based on US Images: A Dual-Center Retrospective Study.

This study aimed to explore the feasibility of using a deep-learning (DL) approach to predict TIL levels in breast cancer (BC) from ultrasound (US) images. A total of 494 breast cancer patients with pathologically confirmed invasive BC from two hospitals were retrospectively enrolled. Of these, 396 patients from hospital 1 were divided into the training cohort (n = 298) and internal validation (IV) cohort (n = 98). Patients from hospital 2 (n = 98) were in the external validation (EV) cohort. TIL levels were confirmed by pathological results. Five different DL models were trained for predicting TIL levels in BC using US images from the training cohort and validated on the IV and EV cohorts. The overall best-performing DL model, the attention-based DenseNet121, achieved an AUC of 0.873, an accuracy of 79.5%, a sensitivity of 90.7%, a specificity of 65.9%, and an F1 score of 0.830 in the EV cohort. In addition, the stratified analysis showed that the DL models had good discrimination performance of TIL levels in each of the molecular subgroups. The DL models based on US images of BC patients hold promise for non-invasively predicting TIL levels and helping with individualized treatment decision-making.

Analysis of angiogenesis-related subtypes of hepatocellular carcinoma and tumor microenvironment infiltration feature in hepatocellular carcinoma.

PURPOSE: Hepatocellular carcinoma (HCC) is a highly vascularized tumor, and angiogenesis plays an important role in its progression. However, the role of angiogenesis in cell infiltration in the tumor microenvironment (TME) remains unclear. METHODS: We evaluated the associations of 35 angiogenesis-related genes (ARGs) with the clinicopathological features of 816 HCC patients. In addition, we assessed the associations between the ARGs and TME cell infiltration. A nomogram was constructed to determine the prognostic value of ARGs for HCC. The ARG score was used to distinguish angiogenic subtypes of HCC, and its usefulness for predicting the prognosis and treatment response of HCC patients was evaluated. RESULTS: We distinguished three ARG clusters differing in terms of TME cell infiltration, immune cell activation status, clinicopathological features, and clinical outcomes. There were significant associations of ARG expression with tumor immunity, the epithelial-mesenchymal transition (EMT), and transforming growth factor-ß expression. An ARG score model was constructed to generate a risk score for each patient based on differentially expressed genes between clusters. Furthermore, a high ARG score was associated with high expression of CTLA-4 and PD-L1/PD-1, and a low Tumor Immune Dysfunction and Exclusion score, indicating the usefulness of the ARG score for selecting patients for immunotherapy. Considering the relationship between ARGs and tumor immunity, immunotherapy combined with vascular-targeted therapy may be the best treatment for HCC. CONCLUSIONS: ARGs play an important role in TME diversity and complexity in HCC patients. The ARG score of HCC predicts TME invasion and can guide immunotherapy.

Integrated Bioinformatics Analysis for the Screening of Hub Genes and Therapeutic Drugs in Hepatocellular Carcinoma.

BACKGROUND: Liver cancer is a major medical problem because of its high morbidity and mortality. Hepatocellular carcinoma (HCC) is the most common type of liver cancer. Currently, the mechanism of HCC is unclear, and the prognosis is poor with limited treatment. OBJECTIVE: The purpose of this study is to identify hub genes and potential therapeutic drugs for HCC. METHODS: We used the GEO2R algorithm to analyze the differential expression of each gene in 4 gene expression profiles (GSE101685, GSE62232, GSE46408, and GSE45627) between HCC and normal hepatic tissues. Next, we screened out the differentially expressed genes (DEGs) by corresponding calculation data according to adjusted P-value < 0.05 and | log fold change (FC) | > 1.0. Subsequently, we used the DAVID software to analyze the DEGs by GO and KEGG enrichment analysis. Then, we carried out the protein-protein interaction (PPI) network analysis of DEGs using the STRING tool, and the PPI network was constructed by Cytoscape software. MCODE plugin was used for module analysis, and the hub genes were screened out by the Cyto- Hubba plugin. Meanwhile, we used The Kaplan-Meier plotter, GEPIA2 and HPA databases to exert survival analysis and verify the expression alternation of hub genes. Furthermore, we used ENCORI, TargetScan, miRDB and miRWalk database to predict the upstream regulated miRNA of hub genes and construct a miRNA-hub genes network by Cytoscape software. Finally, we selected potential therapeutic drugs for HCC through DGIdb databases. RESULTS: A total of 415 DEGs were screened in HCC, including 196 up-regulated DEGs and 219 down-regulated DEGs. The results of KEGG pathway analysis suggested that the up-regulated DEGs can regulate the cell cycle, and DNA replication signal pathway, while the down-regulated DEGs were associated with metabolic pathways. In this study, we identified 11 hub genes (AURKA, BUB1B, TOP2A, MAD2L1, CCNA2, CCNB1, BUB1, KIF11, CDK1, CCNB2 and TPX2), which were independent risk factors of HCCand all up-regulated DEGs. We verified the expression difference of hub genes through the GEPIA2 and HPA database, which was consistent with the results of GEO data. We found that those hub genes were mutations in HCC according to the cBioPortal database. Finally, we used the DGIdb database to select 32 potential therapeutic targeting drugs for hub genes. CONCLUSION: In summary, our study provided a new perspective for researching the molecular mechanism of HCC. Hub genes, miRNAs, and candidate drugs provide a new direction for the early diagnosis and treatment of HCC.

Prolonged warm ischemia aggravates hepatic mitochondria damage and apoptosis in DCD liver by regulating Ca2+/CaM/CaMKII signaling pathway.

This study was conducted to investigate the effect of warm ischemia duration on hepatocyte mitochondrial damage after liver transplantation, and confirm the role of CaMKIIγ in this process. Rat donation after cardiac death (DCD) liver transplantation model was established by exposing donor liver to 0 (W0 group), 15 (W15 group), and 30 (W30 group) min warm ischemia. Some rats in W15 group were transfected with CaMKIIγ and CaMKIIγ-shRNA lentivirus. On day 1, 3, and 7 post-transplantation, a series of experiments, including HE staining, TEM observation, ALT and AST measurement, flow cytometry analysis, qRT-PCR, and Western blotting were performed to evaluate the extent of hepatic and mitochondria damage. Within 7 days post-transplantation, prolonged ischemia led to an obvious deterioration of hepatic and mitochondria damage, presenting with a marked increase of apoptotic hepatocytes, ALT and AST levels, cells with low MMP, and AIF and Cyt C expression. CaMKIIγ overexpression caused the significant ultrastructural damage of hepatic cells, increase of cells with low MMP, enhancement of AIF and Cyt C expression, and augmented Ca2+/CaM/CaMKIIγ, while blocking CaMKIIγ showed an opposite result. In conclusion, ischemia duration is proportional to the extent of hepatic mitochondria damage, and CaMKIIγ plays a negative regulatory role in this process by regulating the Ca2+/CaM/CaMKII signaling pathway.