Harmonization of multi‑site diffusion tensor imaging data for cervical and thoracic spinal cord at 1.5 T and 3 T using longitudinal ComBat

Scientific Reports, 2023 · DOI: https://doi.org/10.1038/s41598-023-46465-6 · Published: November 1, 2023

Neuroimaging Research Methodology & Design

Simple Explanation

This study addresses the challenge of combining diffusion tensor imaging (DTI) data from different MRI scanners for the spinal cord. DTI is used to study spinal cord injuries and diseases, but differences in scanners can make it hard to compare data. The researchers used a method called longitudinal ComBat (longComBat) to reduce the differences between data from different scanners. They scanned thirty healthy adults on four different MRI scanners at different field strengths (1.5T and 3T). The study found that longComBat significantly improved the agreement between DTI measurements from different scanners. This suggests that it is possible to combine DTI data from different sites and scanners for spinal cord research and clinical trials.

Study Duration

Not specified

Participants

30 healthy adult volunteers

Evidence Level

Not specified

Key Findings

1
Large variability was found in diffusivity measures across different scanners, particularly in the upper thoracic region.
2
Harmonized metrics using longComBat showed greatly improved agreement between scanners for all DTI metrics.
3
The use of longComBat resulted in a substantial reduction in variance between datasets acquired from different scanners.

Research Summary

This study examines the variability of DTI of the spinal cord between MR scanners of differing field strengths, hardware capabilities, and manufacturers by scanning a sample of thirty healthy adult subjects on four MR scanners at field strengths of 1.5 T and 3 T. The study demonstrates the efficacy of longComBat in decreasing scanner effects on the data from different scanners and field strengths. The study concludes that harmonization of the diffusion data in the human spinal cord is a critical step towards opportunities for longitudinal and multisite clinical research and clinical trials.

Practical Implications

Improved Multi-site Research

The harmonization method allows for more reliable combination of data from different sites in clinical trials and research.

Longitudinal Study Reliability

Enhances the reliability of longitudinal studies where patients may be scanned on different machines over time.

Clinical Application

Facilitates the use of DTI as a biomarker in clinical settings by reducing the impact of scanner-specific variations.

Study Limitations

1
The use of different field strengths and sequence parameters is a limitation of the study in terms of strict reliability.
2
Subject motion was present in a number of scans which confounded the unavoidable problem of physiologic motion from cardiac and respiratory cycles.
3
The longComBat method requires a minimum of two scans per scanner.

Your Feedback

Was this summary helpful?

Back to Neuroimaging