Background: Early-stage diagnosis and treatment can improve survival rates of liver cancer patients. Dynamic contrast-enhanced MRI provides the most comprehensive information for differential diagnosis of liver tumors. However, MRI diagnosis is affected by subjective experience, so deep learning may supply a new diagnostic strategy. We used convolutional neural networks (CNNs) to develop a deep learning system (DLS) to classify liver tumors based on enhanced MR images, unenhanced MR images, and clinical data including text and laboratory test results. Methods: Using data from 1,210 patients with liver tumors (N = 31,608 images), we trained CNNs to get seven-way classifiers, binary classifiers, and three-way malignancyclassifiers (Model A-Model G). Models were validated in an external independent extended cohort of 201 patients (N = 6,816 images). The area under receiver operating characteristic (ROC) curve (AUC) were compared across different models. We also compared the sensitivity and specificity of models with the performance of three experienced radiologists. Results: Deep learning achieves a performance on par with three experienced radiologists on classifying liver tumors in seven categories. Using only unenhanced images, CNN performs well in distinguishing malignant from benign liver tumors (AUC, 0.946; 95% CI 0.914-0.979 vs. 0.951; 0.919-0.982, P = 0.664). New CNN combining unenhanced images with clinical data greatly improved the performance of classifying malignancies as hepatocellular carcinoma (AUC, 0.985; 95% CI 0.960-1.000), metastatic tumors (0.998; 0.989-1.000), and other primary malignancies (0.963; 0.896-1.000), and the agreement with pathology was 91.9%.These models mined diagnostic information in unenhanced images and clinical data by deepneural-network, which were different to previous methods that utilized enhanced images. The sensitivity and specificity of almost every category in these models reached the same high level compared to three experienced radiologists. Zhen et al. Deep Learning for Liver Tumor Diagnosis Conclusion: Trained with data in various acquisition conditions, DLS that integrated these models could be used as an accurate and time-saving assisted-diagnostic strategy for liver tumors in clinical settings, even in the absence of contrast agents. DLS therefore has the potential to avoid contrast-related side effects and reduce economic costs associated with current standard MRI inspection practices for liver tumor patients.
Human papillomavirus (HPV) is an etiological risk factor for a subset of head and neck squamous cell carcinoma (HNSCC). HPV+ HNSCC is significant more radiosensitive than HPV-HNSCC, but the underlying mechanism is still unknown. Tumor microenvironment can affect tumor response to radiation therapy. Cancer secreted exosomes are emerging as crosstalk mediators between tumor cells and the tumor microenvironment. The main objectives of this study were to determine the role of HPV+ HNSCC-derived exosomes in increased radiation sensitivity. Here, we found that exosomes derived from HPV+ HNSCC cells activate macrophages into the M1 phenotype, which then increases the radiosensitivity of HNSCC cells. miR-9 was enriched in exosomes released from HPV+ HNSCC cells and it could be transported to macrophages, leading to altered cellular functions. Overexpression of miR-9 in macrophages induced polarization into the M1 phenotype via downregulation of PPARδ. Increased radiosensitivity was observed for HNSCC cells co-cultured with macrophages in which miR-9 was upregulated or treated with M1 macrophages. These observations suggest that HPV+ HNSCC cells secrete miR-9-rich exosomes, which then polarize macrophages into M1 phenotype and lead to increased radiosensitivity of HNSCC cells. Hence, miR-9 may be a potential treatment strategy for HNSCC..
A family of small non-coding RNAs, ~22 nt in length, known as microRNAs (miRNAs), regulating ~30% of all human gene expression, have been reported to be involved in the pathogenesis of a number of types of cancers, including laryngeal squamous cell carcinoma (LSCC). In the current study, miR-34a and miR-34c were observed to be downregulated in human LSCC tissues. Ectopic expression of miR-34a and miR-34c in Hep-2 cells significantly induced the cell proliferation and migration ability in vitro. UDP-N-acetyl-α-D-galactosamine:polypeptide-N-acetylgalactosaminyltransferase 7 (GALNT7), whose expression is negatively regulated by miR-34a and miR-34c in Hep-2 cells, is confirmed to be a novel direct target gene of miR-34a and miR-34c. In conclusion, the current results suggest that miR-34a and miR-34c may function as tumor suppressors in LSCC through downregulation of GALNT7. The study of miR-34a, miR-34c and its novel target, GALNT7, may serve as novel potential makers for LSCC therapy.
Laryngeal squamous cell carcinoma (LSCC) is a highly malignant tumor originated from respiratory system. Although there have been many improvements in therapy until now, reducing the high mortality remains difficult. Understanding the cellular heterogeneity of LSCC could contribute to improve this problem. Single-cell RNA sequencing was applied to dissect the cell composition and molecular characteristics of LSCC tissues. Immunohistochemistry staining of the LSCC tissues was performed to identify the spatial location of tumor cells. Survival analysis of marker genes was executed in The Cancer Genome Atlas to verify the correlation between each cell clusters and patients' prognosis. The LSCC tissue cells were finely grouped into various clusters, including tumor cells, immune cells, epithelial cells, fibroblasts and endothelial cells. Notably, in tumor cells, keratinocyte-like cells were in the core of tumor while malignant proliferating cells were located at the tumor edge. The malignant proliferating cells were correlated with poor prognosis. In summary, this is the first study to delineate a landscape of the LSCC intratumor heterogeneity. Our work might help researchers have a better understanding for tumor progression.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.