TORTOISEV4: Reimagining the NIH diffusion MRI processing pipeline

Area of Science:

• Neuroimaging
• Medical Physics
• Biomedical Engineering

Background:

• Diffusion MRI (dMRI) data are prone to various artifacts, including low signal-to-noise ratio (SNR), motion, and distortions.
• Accurate quantitative analysis of dMRI data necessitates appropriate pre-processing before model fitting.
• Evolving dMRI acquisition techniques and large-scale multi-site studies demand advanced processing tools.

Purpose of the Study:

• To present the redesigned and enhanced TORTOISE software for dMRI data processing.
• To address the need for fast, robust, and adaptable dMRI processing pipelines.
• To provide summary reporting capabilities for identifying problematic dMRI data.

Main Methods:

• Redesign and enrichment of the TORTOISE software ensemble.
• Adaptation of the software to handle diverse artifacts and distortions.
• Implementation of summary reporting features for data quality assessment.

Main Results:

• TORTOISE has been significantly improved to meet the demands of modern dMRI studies.
• The enhanced software is adaptable and capable of handling a variety of artifacts.
• The tool facilitates processing for large, multi-site studies, including challenging populations.

Conclusions:

• The enhanced TORTOISE software provides a robust solution for dMRI pre-processing.
• It addresses the challenges posed by complex artifacts and large-scale studies.
• TORTOISE supports accurate quantitative analysis in diverse neuroimaging research settings.