Pseudo-nuclear staining of cells by deep learning improves the accuracy of automated cell counting in a label-free cellular population.

Deep learning has emerged as a breakthrough tool for the segmentation of images without supporting human experts. Here, we propose an automated approach that uses deep learning to generate pseudo-nuclear staining of cells from phase contrast images. Our proposed approach, which has the feature to generate pseudo-nuclear stained images by simple DNN, showed remarkable advantages over existing approaches in the cell-detection and the detection of the relative position of cells for various cell densities, as well as in counting the exact cell numbers. Pseudo-nuclear staining of cells by deep learning will improve the accuracy of automated cell counting in a label-free cellular population on phase contrast images.

Development of an automated two pronuclei detection system on time-lapse embryo images using deep learning techniques.

PURPOSE: To establish an automated pronuclei determination system by analysis using deep learning technology which is able to effectively learn with limited amount of supervised data. METHODS: An algorithm was developed by explicitly incorporating human observation where the outline around pronuclei is being observed in determining the number of pronuclei. Supervised data were selected from the time-lapse images of 300 pronuclear stage embryos per class (total 900 embryos) clearly classified by embryologists as 0PN, 1PN, and 2PN. One-hundred embryos per class (a total of 300 embryos) were used for verification data. The verification data were evaluated for the performance of detection in the number of pronuclei by regarding the results consistent with the judgment of the embryologists as correct answers. RESULTS: The sensitivity rates of 0PN, 1PN, and 2PN were 99%, 82%, and 99%, respectively, and the overlapping 2PN being difficult to determine by microscopic observation alone could also be appropriately assessed. CONCLUSIONS: This study enabled the establishment of the automated pronuclei determination system with the precision almost equivalent to highly skilled embryologists.

Factors associated with longitudinal changes in serum concentrations of Mac-2 binding protein: A prospective 3-year observational study.

BACKGROUND AND AIMS: Mac-2 binding protein (M2BP) plays an important role in cell adhesion. In a recent cross-sectional study we reported that serum M2BP concentrations may reflect silent atherosclerosis. The aim of the present prospective follow-up study was to investigate possible relationships between changes in concentrations of M2BP and other factors over a >3-year period. METHODS AND RESULTS: The present study enrolled subjects who visited Enshu hospital from 2014 to 2015 for a periodic physical check-up and then attended for another physical check-up after >3 years (n = 174). Factors affecting changes in M2BP concentrations were investigated at both baseline and follow-up. Subjects with liver dysfunction, a history of hepatic disease, malignant neoplasm, or cardiovascular events at baseline were excluded. Univariate and multivariate regression analyses showed that changes in serum M2BP concentrations during the follow-up period were significantly associated with changes in low-density lipoprotein cholesterol (LDL-C), triglyceride, and oxidative stress marker derivatives of reactive oxygen metabolites (d-ROM) concentrations. Moreover, the increase in LDL-C was significantly greater in subjects in whom M2BP concentrations increased during the follow-up period. Logistic regression analysis with an endpoint of increased M2BP revealed that increased LDL-C was an independent determinant of an increase in M2BP during the follow-up period. CONCLUSION: During the observation period of >3 years, serum M2BP concentrations were increased in subjects who also exhibited increases in levels of metabolic parameters, especially LDL-C, and the oxidative stress marker d-ROM. These results support that serum M2BP reflects one of the contributors to the progression of silent atherosclerosis.

Clinical Evaluation of Hepatocarcinogenesis and Outcome Using a Novel Glycobiomarker Wisteria floribunda Agglutinin-Positive Mac-2 Binding Protein (WFA+-M2BP) in Chronic Hepatitis C with Advanced Fibrosis.

This study aimed to assess the association between the serum glycobiomarker Wisteria floribunda agglutinin-positive Mac-2 binding protein (WFA+-M2BP) for liver fibrosis and outcomes and carcinogenesis of hepatocellular carcinoma (HCC) in chronic hepatitis C (CHC) patients with advanced fibrosis. Serum WFA+-M2BP levels were measured in 128 consecutive CHC patients including 49 with HCC histopathologically diagnosed with advanced fibrosis (44 with fibrosis stage F3 and 84 with fibrosis stage F4) in our hospital. The median WFA+-M2BP level was significantly higher in F4 than in F3 patients (6.9 vs. 2.3 cutoff index [COI], respectively; p<0.001). The difference in WFA+-M2BP levels between patients with and without HCC was not significant. The respective 5-/8-yr survival rates of patients without HCC at enrollment with high (≥4 COI, n=39), intermediate (1-4 COI, n=33), and low WFA+-M2BP (<1 COI, n=7) levels were 78%/48%, 100%/82%, and 100%/100%, respectively. The differences in survival rates between groups were significant (p=0.0041). Patients with high WFA+-M2BP levels had a significantly higher incidence of HCC than those with low WFA+-M2BP levels (p=0.0019). Cumulative 5-yr carcinogenesis rates in patients with high, intermediate, and low WFA+-M2BP levels were 48.7%, 16.9%, and 0%, respectively; the differences between groups were significant (p=0.002). Serum WFA+-M2BP levels might allow the prediction of carcinogenesis and outcome in CHC patients with advanced fibrosis.

Effect of water-immersion restraint stress on tryptophan catabolism through the kynurenine pathway in rat tissues.

The aim of this study was to clarify the effect of water-immersion restraint stress (WIRS) on tryptophan (Trp) catabolism through the kynurenine (Kyn) pathway in rat tissues. The tissues of rats subjected to 6 h of WIRS (+WIRS) had increased tryptophan 2,3-dioxygenase (TDO) and indoleamine 2,3-dioxygenase (IDO) activities and increased TDO and IDO1 (one of two IDO isozymes in mammals) mRNA expression levels, with decreased Trp and increased Kyn contents in the liver. +WIRS rats had unchanged TDO and IDO activities in the kidney, decreased TDO activity and unchanged IDO activity in the brain, and unchanged IDO activity in the lung and spleen, with increased Kyn content in all of these tissues. Pretreatment of stressed rats with RU486, a glucocorticoid antagonist, attenuated the increased TOD activity, but not the increased IDO activity, with partial recoveries of the decreased Trp and increased Kyn contents in the liver. These results indicate that WIRS enhances hepatic Trp catabolism by inducing both IDO1 and TDO in rats.

Serum levels of Mac-2 binding protein increase with cardiovascular risk and reflect silent atherosclerosis.

BACKGROUND AND AIMS: Mac-2 binding protein (M2BP) was reported to be a useful biomarker for liver fibrosis and malignant tumors. We hypothesized that expression of M2BP might also change in the process of atherosclerosis. METHODS: This study included subjects who visited our hospital for a physical checkup. RESULTS: The M2BP levels in subjects with hypertension, dyslipidemia, or abnormal glucose metabolism were higher than those in subjects without such risk factors. Moreover, the M2BP levels were associated with severity of cardiovascular risk. Subdivision of M2BP levels into quartiles revealed that M2BP was significantly associated with reactive oxygen metabolites, central systolic blood pressure, and radial augmentation index (AI). Logistic regression analysis with the endpoint of high radial AI (above mean value) showed that high radial AI was independently associated with high M2BP. CONCLUSIONS: Although the spectrum was narrow as compared to that in cases of hepatic fibrosis, serum M2BP may reflect silent atherosclerosis in apparently healthy subjects.

Virological characteristics of hepatitis B genotype G/A2 recombination virus in Japan.

AIM: We identified four cases of infection with hepatitis B virus genotype G and A2 recombinant (HBV/G/A2) strains, which were initially overlooked by enzyme immunoassay-based genotyping. The patients were all men who have sex with men (MSM) and inhabited several metropolitan areas of Japan, suggesting that the recombinant strains may be circulating among high-risk groups such as MSM. Here, we investigated the genomic structure and virological properties of the HBV/G/A2 strains. METHODS: Complete genome sequences of the isolates were determined and phylogenetically analyzed. Replication efficiency of HBV/G/A2 was investigated by transfecting plasmids containing 1.24-fold viral genome. The in vivo viral kinetics of HBV/G/A2 were investigated using chimeric mice with humanized livers. RESULTS: Phylogenetic analysis revealed that the four strains were almost identical (>99.7% homologous). The preS2/S region of these strains was highly homologous to that of genotype A2 and the remaining region was almost identical to that of genotype G, reflecting inter-genotypic recombination. Interestingly, in all four cases, genotype A was co-infected as a minor population. In vitro analysis revealed that HBV/G/A2 had a low replication rate. Although detectable viremia was not measurable following the inoculation of HBV/G/A2 into chimeric mice, subsequent superinfection of HBV genotype A greatly enhanced HBV/G/A2 replication and viral spread. CONCLUSION: We found that four cases of HBV/G/A2 recombinant among MSM patients in the metropolitan areas of Japan, and HBV/A co-infections are required for its efficient replication. High-risk groups such as MSM should be carefully tested for infection of genotype G-derived variants.