Quantification of relative neurite tortuosity using Fourier transforms

Area of Science:

• Neuroscience
• Biomedical Imaging
• Ophthalmology

Background:

• Nerve fiber tortuosity is a key indicator for various ocular and systemic diseases.
• Quantifying tortuosity in dense nerve networks is challenging with current methods requiring manual segmentation or scoring.

Purpose of the Study:

• To develop a novel method for accurately quantifying nerve fiber tortuosity.
• To enable tortuosity analysis without complete nerve network segmentation.
• To adapt the method for multi-scale analysis.

Main Methods:

• Utilized Fourier transforms of segmented nerve masks to quantify directional coherence.
• Adapted the method to analyze tortuosity at different length and size scales.
• Applied the method to simulated data and murine corneal neurite images.

Main Results:

• Accurately quantified neurite tortuosity in simulated datasets across multiple scales.
• Successfully distinguished scale-variant and scale-invariant tortuosity.
• Identified known differences in corneal neurite tortuosity between central and peripheral regions.

Conclusions:

• Introduced a novel Fourier transform-based method for neurite tortuosity quantification.
• The method works with incompletely segmented neurites and at multiple scales.
• Eliminates the need for manual training or curation, enabling rapid and accurate measurements for disease evaluation.