Development of an AI system for accurately diagnose hepatocellular carcinoma from computed tomography imaging data.

BACKGROUND AND AIMS: Computed tomography (CT) scan is frequently used to detect hepatocellular carcinoma (HCC) in routine clinical practice. The aim of this study is to develop a deep-learning AI system to improve the diagnostic accuracy of HCC by analysing liver CT imaging data. METHODS: We developed a deep-learning AI system by training on CT images from 7512 patients at Henan Provincial Peoples' Hospital. Its performance was validated on one internal test set (Henan Provincial Peoples' Hospital, n = 385) and one external test set (Henan Provincial Cancer Hospital, n = 556). The area under the receiver-operating characteristic curve (AUROC) was used as the primary classification metric. Accuracy, sensitivity, specificity, precision, negative predictive value and F1 metric were used to measure the performance of AI systems and radiologists. RESULTS: AI system achieved high performance in identifying HCC patients, with AUROC of 0.887 (95% CI 0.855-0.919) on the internal test set and 0.883 (95% CI 0.855-0.911) on the external test set. For internal test set, accuracy was 81.0% (76.8-84.8%), sensitivity was 78.4% (72.4-83.7%), specificity was 84.4% (78.0-89.6%) and F1 (harmonic average of precision and recall rate) was 0.824. For external test set, accuracy was 81.3% (77.8-84.5%), sensitivity was 89.4% (85.0-92.8%), specificity was 74.0% (68.5-78.9%) and F1 was 0.819. Compared with radiologists, AI system achieved comparable accuracy and F1 metric on internal test set (0.853 versus 0.818, P = 0.107; 0.863 vs. 0.824, P = 0.082) and external test set (0.805 vs. 0.793, P = 0.663; 0.810 vs. 0.814, P = 0.866). The predicted HCC risk scores by AI system in HCC patients with multiple tumours and high fibrosis stage were higher than those with solitary tumour and low fibrosis stage (tumour number: 0.197 vs. 0.138, P = 0.006; fibrosis stage: 0.183 vs. 0.127, P < 0.001). Radiologists' review showed that the accuracy of saliency heatmaps predicted by algorithms was 92.1% (95% CI: 89.2-95.0%). CONCLUSIONS: AI system achieved high performance in the detection of HCC compared with a group of specialised radiologists. Further investigation by prospective clinical trials was necessitated to verify this model.

Two-Pronged Treatment of Hemoperitoneum and Abnormal Uterine Bleeding in an Adolescent Girl With Congenital Fibrinogen Deficiency.

Introduction: If a woman suffers from congenital fibrinogen deficiency (CFD), she might undergo hypermenorrhea monthly and possibly to suffer from hemoperitoneum due to a ruptured follicle or corpus luteal cyst every month, which seriously threaten her health and quality of life. Here, we creatively used a combination of the levonorgestrel-releasing intrauterine system (LNG-IUS) and the combined oral contraceptives (COC) for a girl with CFD. Case presentation: A 14-year-old girl presented with no obvious cause of persistent and severe lower abdominal pain that began 8 h prior. After examination, she was diagnosed as hemoperitoneum. CFD was diagnosed when she was 2 years old and she had two hospitalizations due to hemorrhagic anemia caused by menorrhagia. Therefore, after successful conservative treatment of hemoperitoneum, a combination of LNG-IUS and COC was used for the long-term conservative management of hypermenorrhea and hemoperitoneum. During the half-year of follow-up, she had hypomenorrhea without hemoperitoneum. Conclusions: To the best of our knowledge, this was the first patient treated with such a procedure in the literature, and we recommend every woman with CFD at puberty or reproductive age receives this two-pronged treatment.

High-concentration glucose enhances invasion in invasive ductal breast carcinoma by promoting Glut1/MMP2/MMP9 axis expression.

Type 2 diabetes mellitus (T2DM) has been considered to be a risk factor for numerous human cancers. Hyperglycemia is one of the most direct internal environmental changes for patients with T2DM. Increasing evidence reveals that a high concentration of glucose can promote tumor progression, while its role for migration and invasion of invasive ductal breast carcinoma (IDBC) cells remains unclear. In the present study, it was demonstrated that IDBC patients with T2DM suffered an increased tumor size and more frequent lymphatic and distant metastasis compared with those without T2DM (P<0.05). MCF-7 breast carcinoma cells, which were cultured in a high glucose concentration medium (25.00 mM), exhibited increased invasion (P<0.05). In addition, the expression of glucose transporters (Gluts), matrix metalloproteinase 2 (MMP2) and matrix metalloproteinase 9 (MMP9) in IDBC tissues with T2DM was significantly higher compared to those without T2DM. Downregulation of glucose transporter 1 (Glut1) by small interfering RNA may markedly suppress MCF-7 cell invasion as well as the expression of MMP2 and MMP9. These results suggest that T2DM can affect the malignant features of tumors in IDBC. The high glucose concentration in the tumor microenvironment may enhance IDBC invasion via upregulating Glut1/MMP2/MMP9 axis expression.

Non-canonical NF-κB Plays a Pivotal Role in Non-Hodgkin's Lymphoma.

This study tried to dissect BAFF/BAFF-R-mediated non-canonical NF-κB signaling in the drug resistance of B-cell non-Hodgkin's lymphoma. FQ-PCR was employ to determine the mRNA expression of BAFF, BAFF-R, Bcl-xL, and RIPK2 in the circulating blood of B-cell non-Hodgkin's lymphoma patients. Further correlation studies were performed with the gene expression in the circulating blood and tumor tissue. MTT assay as used to determine BAFF's role on lymphoma cell proliferation. Western blot was employed to determine protein expression after BAFF stimulation. The mRNA expression of BAFF, BAFF-R, Bcl-xL, and RIPK2 in the circulating blood of the resistant group was higher than that of the non-resistant group, which was statistically significant. The mRNA expression of the target genes was positively correlated. The mRNA expression was positively correlated with disease progression, which was statistically significant. The Bcl-xL mRNA expression in the resistant group was relatively higher than that in the non-resistant group, which was also statistically significant. However, the mRNA expression of other genes only showed increased tendency compared with non-resistant group. There was no significant change between target genes in different tumor tissues. With increased BAFF concentration and prolonged exposure, the proliferation of the tumor cells increased significantly, which was statistically significant. Western blot showed the expression of BAFF, BAFF-R, Bcl-xL, and RIPK2 all increased with increased BAFF concentration, which was also statistically significant. In B-cell, non-Hodgkin's lymphoma, BAFF may activate non-canonical NF-κB signaling to regulate drug resistance.

[Dynamic changes of adenomatous polyposis coli protein and glycogen synthase kinase 3beta in the repair of the injured airway epithelial cells in smoking mice].

To investigate the roles of adenomatous polyposis coli (APC) protein and glycogen synthase kinase 3beta (GSK3beta) of smoking murine model in the repair of the injured airway epithelial cells (AECs) in different stages, 30 male Kun-Ming mice were randomly divided into two groups, the control group and the smoking group. There were 24 mice in smoking group, and 6 animals were separately killed at the end of the 1st, 4th, 8th and 12th week after smoking. Then the following tests were undertaken: (1) HE staining of lung section to observe the morphological changes of the bronchi in the smoking mice. (2) Immunohistochemical staining of APC protein and GSK3beta in the AECs. (3) Western blot was used to detect the levels of APC protein, GSK3beta and phosphorated GSK3beta (p-GSK3beta) in pulmonary tissue. (4) Observing the localizations of APC protein and GSK3beta in the AECs by immunofluorescence technique. The results showed: (1) AECs showed changes of predominant injury (1-, 4-week), repair (8-week) and reinjury (12-week) along with smoking time prolonged. The experimental results indicated that the model of smoking mice was duplicated successfully. (2) Immunohistochemical results showed that the expression of APC protein in the AECs increased after 1-week smoking (0.458 +/- 0.062 vs 0.399 +/- 0.060, P< 0.05 vs control), but was significantly decreased at the end of the 4th week (0.339+/- 0.056, P<0.01 vs control) and increased at the end of the 8th and 12th week (0.387 +/- 0.041, 0.378 +/- 0.037, P<0.05 vs 4-week). The expression of GSK3beta in the AECs of smoking mice obviously decreased (P<0.01 or P<0.05 vs control). (3) Western blot showed that the expressions of APC protein and GSK3beta in lung tissue were consistent with the results of immunohistochemistry; and the levels of p-GSK3beta in all smoking models were higher than that in control. (4) The results of immunofluorescence showed that APC protein was localized mainly near the regions of epithelial cell membrane at the end of the 1st and 8th week after smoking, which were dissimilar with the localization in control, and this change was not seen in the location of GSK3beta. Taken together, these results demonstrate that the expressions and localizations of APC protein, GSK3beta and the activity of GSK3beta are dynamically changed in the AECs with experimental smoking injury at different phases, suggesting that APC protein and GSK3beta may be involved in the regulation of migration and proliferation of AECs, and play an important role in the process of repair of airway epithelium injury.