QuNex-An integrative platform for reproducible neuroimaging analytics

Area of Science:

• Neuroscience
• Computational Biology
• Bioinformatics

Background:

• Neuroimaging research faces challenges integrating diverse tools and data across modalities and species.
• Siloed approaches limit reproducibility, data organization, and software interoperability in neuroimaging studies.
• Existing methods hinder scalable and consistent analysis of neural mechanisms in health and disease.

Purpose of the Study:

• To develop a platform for consistent, end-to-end neuroimaging data processing and analytics.
• To address challenges in method integration, reproducibility, and interoperability in the neuroimaging field.
• To create a cohesive translational platform for integrating human and non-human neuroimaging workflows.

Main Methods:

• Developed Quantitative Neuroimaging Environment & Toolbox (QuNex) for reproducible deployment of custom workflows.
• Implemented an extension framework with an SDK for integrating community-developed neuroimaging software.
• Enabled high-throughput, parallel processing in local and cloud high-performance compute environments.

Main Results:

• QuNex successfully processed over 10,000 scans across multiple neuroimaging consortia and clinical datasets.
• The platform demonstrated seamless integration of multi-modal neuroimaging software.
• Facilitated the integration of human and non-human workflows within a cohesive translational platform.

Conclusions:

• QuNex significantly impacts neuroimaging method integration across acquisition approaches, pipelines, datasets, computational environments, and species.
• The platform enables more rapid, scalable, and reproducible advancements in neuroimaging technology.
• QuNex fosters collaborative and standardized research in neuroscience for understanding health and disease.