ResBCDU-Net: A Deep Learning Framework for Lung CT Image Segmentation.

Lung CT image segmentation is a key process in many applications such as lung cancer detection. It is considered a challenging problem due to existing similar image densities in the pulmonary structures, different types of scanners, and scanning protocols. Most of the current semi-automatic segmentation methods rely on human factors therefore it might suffer from lack of accuracy. Another shortcoming of these methods is their high false-positive rate. In recent years, several approaches, based on a deep learning framework, have been effectively applied in medical image segmentation. Among existing deep neural networks, the U-Net has provided great success in this field. In this paper, we propose a deep neural network architecture to perform an automatic lung CT image segmentation process. In the proposed method, several extensive preprocessing techniques are applied to raw CT images. Then, ground truths corresponding to these images are extracted via some morphological operations and manual reforms. Finally, all the prepared images with the corresponding ground truth are fed into a modified U-Net in which the encoder is replaced with a pre-trained ResNet-34 network (referred to as Res BCDU-Net). In the architecture, we employ BConvLSTM (Bidirectional Convolutional Long Short-term Memory)as an advanced integrator module instead of simple traditional concatenators. This is to merge the extracted feature maps of the corresponding contracting path into the previous expansion of the up-convolutional layer. Finally, a densely connected convolutional layer is utilized for the contracting path. The results of our extensive experiments on lung CT images (LIDC-IDRI database) confirm the effectiveness of the proposed method where a dice coefficient index of 97.31% is achieved.

MHiC, an integrated user-friendly tool for the identification and visualization of significant interactions in Hi-C data.

BACKGROUND: Hi-C is a molecular biology technique to understand the genome spatial structure. However, data obtained from Hi-C experiments is biased. Therefore, several methods have been developed to model Hi-C data and identify significant interactions. Each method receives its own Hi-C data structure and only work on specific operating systems. RESULTS: We introduce MHiC (Multi-function Hi-C data analysis tool), a tool to identify and visualize statistically signifiant interactions from Hi-C data. The MHiC tool (i) works on different operating systems, (ii) accepts various Hi-C data structures from different Hi-C analysis tools such as HiCUP or HiC-Pro, (iii) identify significant Hi-C interactions with GOTHiC, HiCNorm and Fit-Hi-C methods and (iv) visualizes interactions in Arc or Heatmap diagram. MHiC is an open-source tool which is freely available for download on https://github.com/MHi-C. CONCLUSIONS: MHiC is an integrated tool for the analysis of high-throughput chromosome conformation capture (Hi-C) data.

Spatio-temporal deep learning for EEG-fNIRS brain computer interface.

In this paper the classification of motor imagery brain signals is addressed. The innovative idea is to use both temporal and spatial knowledge of the input data to increase the performance. Definitely, the electrode locations on the scalp is as important as the acquired temporal signals from every individual electrode. In order to incorporate this knowledge, a deep neural network is employed in this work. Both motor-imagery EEG and bi-modal EEG-fNIRS datasets were used for this purpose. The results are compared for different scenarios and using different methods. The achieved results are promising and imply that combining both temporal and spatial information of the brain signals could be really effective and increases the performance.

The effect of methenamine on vascular development: Experimental investigation using in vivo and insilico methods.

BACKGROUND: Methenamine is a worldwide antibacterial agent for urinary system infections in human and animals. The effect of methenamine consumption during early phase of pregnancy is not fully clarified in previous studies. Vascular development is the essential part of the early embryonic growth. OBJECTIVE: In this study, we used chicken chorioallantoic membrane to evaluate the effects of methenamine administration on angiogenesis process as a model. MATERIALS AND METHODS: In this experimental study, 20 Ross 308 eggs (mean weight 55 ± 4) were incubated. The eggs were divided into two equal groups (n = 10/each). In the first group, methenamine (150 mg/kg egg weight) was injected on the shell membrane, and in the second group (control group) phosphate-buffered salineas injected. Methenamine was inoculated at 96 and 120 hrafter incubation; 24 hrafter the last inoculation, the eggs were removed and the egg's shell was incised. Then, the development of vascular network and vascular endothelial growth factor Aexpression was evaluated. RESULTS: Angiogenesis was significantly decreased after methenamine treatment. The indexes such as areas containing vessels, the vessels' length, the percentage of angiogenesis developing areas, and vascular complexity in the treatment group receiving methenamine were significantly reduced compared to the control group. Vascular endothelial growth factor Aexpression was suppressed in the methenamine treated group. CONCLUSION: According to the achieved results, it was defined that methenamine could have an inhibitory effect on the growth and development procedures of extraembryonic vasculature.