About

Northwestern University Research Image Processing System (NURIPS) is a web-based data repository and processing system. The goal of the system is to enable a seamless workflow from data storage to results with support for any imaging modality. Imaging data is stored in our data repository from multiple sources including CTI and external datasets. Users can select the entire project or individual subjects for processing. Once a pipeline is specified, the data is transferred to our high performance compute resources (Quest or our NeuroCluster), processed and results are transferred back for longterm storage. We are also integrating AI workflows. Pipelines can be designated as Public or Private and all labs are able to develop on our platform. Each pipeline is optimized using OpenMP or GPU acceleration in order to maximize throughput of data processing.

Features

Containers

Our system uses Singularity images in order to make the system more stable and enable robust version control. Singularity is a container system similar to Docker with enhanced security to run in HPC environments such as Quest

NeuroCluster

In order to improve throughput of our system, we built a dedicated 5 node server system with room for future expansion to process data alongside the Quest high performance computer. Each node has 64 cores, 256 GB ram and an NVIDIA A30 GPU and attached to our 400 TB storage system

Protocol Validation

Validate basic imaging parameters upon data upload.

Northwestern University Research Image Processing System

ABOUT

Features

LOCATION

Laboratory