Search

Conventional deep learning (DL) algorithm requires full supervision of annotating the region of interest (ROI) that is laborious and often biased. We aimed to develop a weakly-supervised DL algorithm that diagnosis breast cancer at ultrasound without image annotation. Weakly-supervised DL algorithms were implemented with three networks (VGG16, ResNet34, and GoogLeNet) and trained using 1000 unannotated US images (500 benign and 500 malignant masses). Two sets of 200 images (100 benign and 100 malignant masses) were used for internal and external validation sets. For comparison with fully-supervised algorithms, ROI annotation was performed manually and automatically. Diagnostic performances were calculated as the area under the receiver operating characteristic curve (AUC). Using the class activation map, we determined how accurately the weakly-supervised DL algorithms localized the breast masses. For internal validation sets, the weakly-supervised DL algorithms achieved excellent diagnostic performances, with AUC values of 0.92–0.96, which were not statistically different (all Ps > 0.05) from those of fully-supervised DL algorithms with either manual or automated ROI annotation (AUC, 0.92–0.96). For external validation sets, the weakly-supervised DL algorithms achieved AUC values of 0.86–0.90, which were not statistically different (Ps > 0.05) or higher (P = 0.04, VGG16 with automated ROI annotation) from those of fully-supervised DL algorithms (AUC, 0.84–0.92). In internal and external validation sets, weakly-supervised algorithms could localize 100% of malignant masses, except for ResNet34 (98%). The weakly-supervised DL algorithms developed in the present study were feasible for US diagnosis of breast cancer with well-performing localization and differential diagnosis. [ABSTRACT FROM AUTHOR]

Microgravity in space impacts human health. In particular, thyroid cancer, which has a high incidence rate, has been the subject of numerous studies with respect to microgravity. However, most studies have focused on Western follicular thyroid cancer cell lines, while data regarding the effects of microgravity on Asian cell lines are lacking. Therefore, we aimed to investigate the effect of simulated ground-based microgravity on two Korean thyroid cancer cell lines, namely SNU-790 and SNU-80. We found that both cell lines formed multicellular spheroids under simulated microgravity. Gene expression analysis revealed that in SNU-790 cells, histone-related genes were upregulated and microRNA-related genes were downregulated. Meanwhile, in SNU-80 cells, genes related to the cellular response to hypoxia were downregulated. These findings contribute to a better understanding of the effects of microgravity on thyroid cancer cells. Further validation studies and clinical significance analyses are needed to fully understand the implications of these findings. [ABSTRACT FROM AUTHOR]

We study whether deep neural network based algorithm can filter out mammography phantom images that will pass or fail. With 543 phantom images generated from a mammography unit, we created VGG16 based phantom shape scoring models (multi-and binary-class classifiers). Using these models we designed filtering algorithms that can filter failed or passed phantom images. 61 phantom images obtained from two different medical institutions were used for external validation. The performances of the scoring models show an F1-score of 0.69 (95% confidence interval (CI) 0.65, 0.72) for multi-class classifiers and an F1-score of 0.93 (95% CI 0.92, 0.95) and area under the receiver operating characteristic curve of 0.97 (95% CI 0.96, 0.98) for binary-class classifiers. A total of 42 of the 61 phantom images (69%) were filtered by the filtering algorithms without further need for assessment from a human observer. This study demonstrated the potential to reduce the human workload from mammographic phantom interpretation using the deep neural network based algorithm. [ABSTRACT FROM AUTHOR]

Enterococcus faecalis, a gram-positive bacterium, is among the most common nosocomial pathogens due to its limited susceptibility to antibiotics and its reservoir of the genes coding for virulence factors. Bacterial enzymes such as kinases and phosphorylases play important roles in diverse functions of a bacterial cell and, thus, are potential antibacterial drug targets. In Gram-positive bacteria, HPr Kinase/Phosphorylase (HPrK/P), a bifunctional enzyme is involved in the regulation of carbon catabolite repression by phosphorylating/dephosphorylating the histidine-containing phosphocarrier protein (HPr) at Ser46 residue. Deficiencies in HPrK/P function leads to severe defects in bacterial growth. This study aimed at identifying novel inhibitors of E. faecalis HPrK/P from a commercial compound library using structure-based virtual screening. The hit molecules were purchased and their effect on enzyme activity and growth of resistant E. faecalis was evaluated in vitro. Furthermore, docking and molecular dynamics simulations were performed to study the interactions of the hit compounds with HPrK/P. Among the identified hit molecules, two compounds inhibited the phosphorylation of HPr as well as significantly reduced the growth of resistant E. faecalis in vitro. These identified potential HPrK/P inhibitors open new research avenues towards the development of novel antimicrobials against resistant Gram-positive bacteria. [ABSTRACT FROM AUTHOR]

Background: Mini-screws are widely used as temporary anchorages in orthodontic treatment, but have the disadvantage of showing a high failure rate of about 10%. Therefore, orthodontic mini-screws should have high biocompatibility and retention. Previous studies have demonstrated that the retention of mini-screws can be improved by imparting bioactivity to the surface. The method for imparting bioactivity proposed in this paper is to sequentially perform anodization, periodic pre-calcification, and heat treatments with a Ti–6Al–4V ELI alloy mini-screw. Materials and methods: A TiO2 nanotube-structured layer was formed on the surface of the Ti–6Al–4V ELI alloy mini-screw through anodization in which a voltage of 20 V was applied to a glycerol solution containing 20 wt% H2O and 1.4 wt% NH4F for 60 min. Fine granular calcium phosphate precipitates of HA and octacalcium phosphate were generated as clusters on the surface through the cyclic pre-calcification and heat treatments. The cyclic pre-calcification treatment is a process of immersion in a 0.05 M NaH2PO4 solution and a saturated Ca(OH)2 solution at 90 °C for 1 min each. Results: It was confirmed that the densely structured protrusions were precipitated, and Ca and P concentrations, which bind and concentrate endogenous bone morphogenetic proteins, increased on the surface after simulated body fluid (SBF) immersion test. In addition, the removal torque of the mini-screw fixed into rabbit tibias for 4 weeks was measured to be 8.70 ± 2.60 N cm. Conclusions: A noteworthy point in this paper is that the Ca and P concentrations, which provide a scaffold suitable for endogenous bone formation, further increased over time after SBF immersion of the APH group specimens. The other point is that our mini-screws have a significantly higher removal torque compared to untreated mini-screws. These results represent that the mini-screw proposed in this paper can be used as a mini-screw for orthodontics. [ABSTRACT FROM AUTHOR]