Abnormal auditory mismatch responses to sound duration deviants in a neurodevelopmental rat model of schizophrenia

摘要

Recently, high-resolution image and video data are being increased exponentially in the field of life sciences by the advanced imaging technologies. Among them, connectomics images of brain tissues converted nano and micrometer high-resolution into 3D image produce big data characteristics. In order to visualize such connectomics image effectively, there are resource constraints such as memory and disk. In addition, as Web services grow in scale, sudden increase in traffic causes server bottlenecks as well as degradation of server performance due to server overloading prob- lems. It is essential to establish a Web-based platform which allows the connectomics images to be visualized and analyzed in 3D at anytime and anywhere without restriction on the spatial environ- ment. Also, in order to improve the processing time and speed of the entire image, it is necessary to divide into the image pieces and dis- tribute them to the system. In this study, we propose pre-computed pipeline and methodology generating an automated data set and providing an advantage of block storage in a cloud environment, which are eventually utilized for visualization and analysis of con- nectomics images in three dimensions through a Web browser. The 3D image viewer utilizes neuroglancer developed by Google. The pipeline sets the bounding box space of x, y, z axis and divides into 3D chunk units by slicing work for each area. The divided 3D chunk dataset and information files are kept in the block storage of the cloud and the dataset is converted to the neuroglancer format for efficient I/O operations. This system is aimed to build an interactive database for brain connectome convergence research based on a user interface that can be integrated with various analysis modules.