Deep Learning-Based Universal Expert-Level Recognizing Pathological Images of Hepatocellular Carcinoma and Beyond.

Background and Aims: We aim to develop a diagnostic tool for pathological-image classification using transfer learning that can be applied to diverse tumor types. Methods: Microscopic images of liver tissue with and without hepatocellular carcinoma (HCC) were used to train and validate the classification framework based on a convolutional neural network. To evaluate the universal classification performance of the artificial intelligence (AI) framework, histological images from colorectal tissue and the breast were collected. Images for the training and validation sets were obtained from the Xiamen Hospital of Traditional Chinese Medicine, and those for the test set were collected from Zhongshan Hospital Xiamen University. The accuracy, sensitivity, and specificity values for the proposed framework were reported and compared with those of human image interpretation. Results: In the human-machine comparisons, the sensitivity, and specificity for the AI algorithm were 98.0, and 99.0%, whereas for the human experts, the sensitivity ranged between 86.0 and 97.0%, while the specificity ranged between 91.0 and 100%. Based on transfer learning, the accuracies of the AI framework in classifying colorectal carcinoma and breast invasive ductal carcinoma were 96.8 and 96.0%, respectively. Conclusion: The performance of the proposed AI framework in classifying histological images with HCC was comparable to the classification performance achieved by human experts, indicating that extending the proposed AI's application to diagnoses and treatment recommendations is a promising area for future investigation.

Beneficial Effects of Gracillin From Rhizoma Paridis Against Gastric Carcinoma via the Potential TIPE2-Mediated Induction of Endogenous Apoptosis and Inhibition of Migration in BGC823 Cells.

Tumor necrosis factor-α inducible protein-8 (TIPE2), initially recognized as a negative immune regulator, exerts an important role in suppressing the progression of numerous cancers. In our previous investigation, we found that TIPE2 expression displayed a decrease or absence in gastric tumor tissue, and the overexpression of TIPE2 suppressed the growth of gastric cancer tumors and cells, demonstrating that TIPE2 could be a potential medicinal target for gastric cancer treatment. However, it's seldomly reported that several medicinal agents or candidates targeted TIPE2 for treating diseases, including gastric cancer. To identify the candidate targeting TIPE2 to fight against gastric cancer, several extractions from traditional natural medicinal plants with anti-tumor functions were employed to screen the active compounds according to bioassay-guided isolation. Interestingly, gracillin, a component from the ethyl acetate extraction of Rhizoma Paridis, was identified to induce the expression of TIPE2 and inhibit the cell proliferation in gastric cancer BGC-823 cells. Furthermore, the underlying mechanisms that restrain gastric cancer were evaluated by clone formation, EdU staining, flow cytometry, and other assays. Meanwhile, the role of TIPE2 in the anti-tumor effect of gracillin was elucidated via the use of siTIPE2 RNA. It was determined that gracillin could fight against gastric cancer cells by inhibiting the cell proliferation participated by the PI3K/AKT pathway and cell cycle arrest, suppressing the EMT pathway-regulating cell migration, and inducing bcl2-associated mitochondrial apoptosis. Additionally, TIPE2 maybe contribute to the benefits of gracillin. These results of the present study are an important step toward the medicinal development of gracillin, and are also of use in understanding the effect of TIPE2 as a potential tumor target.

Overexpression of miR-19a inhibits colorectal cancer angiogenesis by suppressing KRAS expression.

The microRNA miR-19a is closely related to tumor formation and development and is a key oncogene. Previous studies have demonstrated that miR-19a is upregulated in multiple cancer types, including colorectal cancer (CRC). However, most of these experiments were performed in vitro, and consequently, the mechanisms underlying the effects of miR-19a on CRC are still unclear. Therefore, in the present study, we investigated the role of miR-19a in the development of solid CRC tumors. We generated KRAS 3'UTR-Mut by deleting the predicted binding site for miR-19a in KRAS, and observed that the expression of a reporter gene containing the KRAS 3'UTR in HCT116 cells was suppressed by miR-19a, while that of a reporter gene with mutant KRAS 3'UTR was unaffected by miR-19a. We observed that high miR-19a levels reduced KRAS expression. In the tube formation assay, overexpression of miR-19a exhibited anti-angiogenesis effects, which were rescued by KRAS expression. We established a nude mouse xenograft model to investigate the specific role of miR-19a in solid tumors. The results revealed that the sizes of xenograft tumors and density of blood vessels developed from HCT116 cells expressing miR-19a were smaller/lower compared with those of the control. KRAS and VEGFA levels were also reduced. In conclusion, our results revealed that miR-19a overexpression supressed KRAS expression and angiogenesis in CRC, indicating possibilities of using miR-19a in future therapeutic applications.