Image Acquisition

The Nanozoomer HT Virtual Microscopy system (Hamamatsu/Olympus) is used to digitally image all tissue samples resulting from the pipeline. All scanning is performed using a 20X objective (0.46µm per pixel). Brightfield scanning is used for Nissl, IHC, and HC stained slides. Fluorescence scanning at 12-bit depth per pixel, per color channel, using a tri-pass filter cube (FITC/TX-RED/DAPI), is used for all other samples (Rabies and AAV injections). A Lumen Dynamics X-Cite exacte light source is used to produce the excitation fluorescence. Cropping boxes are individually positioned around each section and 9-12 focus points are placed per section. Image file names are derived from the barcoded slide name.

Data Conversion and Data Storage

Shortly after image acquisition, the whole-slide data are cropped into individual sections and converted into a custom image data structure. Quality control is performed to ensure that the cropping was done correctly. All data are subsequently transferred to a high-performance storage cluster, through an integrated 10G copper network. Daily backups of the data are created. Data are subjected only to lossless compression; we perform lossy compression for purposes of web display, but raw data is retained in uncompressed form.

Data Presentation Pipeline

Following data conversion, individual brain sections are registered to one another using a rigid-body transformation, and then, using a deformable transformation, are registered to a 3-D digital version of the ARA by custom software developed in MATLAB. Registered 2-D images are placed in a standard bounding box and converted into the JPEG2000 (ISO/IEC 15444-1) image format using custom scripts based on the Kakadu toolkit [9].

The project website has been developed using a combination of technologies, and allows for browsing and viewing high resolution images. The JPEG2000 images are served over the web using a modified version of the Djatoka Image Server [10] and a custom image viewer based on the aDORe Djatoka / IIPMooViewer, which allows for fully interactive zoom and pan, supports online adjustment of RGB dynamic range, as well as gamma adjustment. The website also contains educational units based on the image data and integrates with a variety of other online web tools, such as the Brain Architecture Management System (BAMS) [11].

Laboratory Information Management System (LIMS) and Quality Control

Sample metadata and process-related information are stored in an intranet accessible LIMS system. The system is designed specifically for the pipeline and has dedicated sections for each pipeline station, with multiple fields – each encapsulating an important variable of the process. Status of a sample within the pipeline, priority of processing, reagent control, quality control and overall pipeline management is performed via LIMS. Investigator observations and general notes are recorded at each pipeline station. Specimens are first entered into the system at the step of stereotaxic injection.

A formal quality control procedure is employed to assess the quality of materials at different stages. This is in order to: (A) Determine possible need to re-inject, re-coverslip, re-image; (B) Correct or improve the pipeline process in an evolutionary operation (EvOp) methodology and (C) Flag unusable sections or materials to reduce unnecessary post-processing. Major QC stages, including examination of slides immediately after coverslipping and after scanning, are formalized into a number of attributes to be checked to ensure consistency across operators and for record keeping.