Quantitative evaluation of 10 tractography algorithms on a realistic diffusion MR phantom.

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Computational Neuroscience

Background:

• Diffusion MRI tractography is crucial for in vivo white matter fiber mapping in clinical and neuroscience research.
• The optimal selection of diffusion models and tractography algorithms remains unclear, necessitating quantitative comparisons.

Purpose of the Study:

• To quantitatively evaluate and compare the performance of various diffusion models and tractography algorithms.
• To understand the strengths and weaknesses of different fiber reconstruction methods under varying imaging parameters.

Main Methods:

• Utilized a common dataset with known ground truth and a reproducible methodology for evaluation.
• Conducted a public contest (Fiber Cup) releasing the dataset (excluding ground truth) to assess 10 fiber reconstruction methods.

Main Results:

• For high SNR datasets, diffusion models like orientation distribution functions accurately model fiber distribution with streamline tractography.
• For medium to low SNR datasets, incorporating spatial smoothness priors improves fiber distribution modeling and tractography outcomes.

Conclusions:

• The choice of diffusion MRI tractography method should be guided by the dataset's SNR.
• Publicly available datasets and methodologies facilitate ongoing comparison and development of tractography techniques.