Abstract:

BACKGROUND: Studies have shown that pathological change of substantia nigra (SN) is the main reason for Parkinson's disease (PD). From the medical imaging, spatial location, size and 3D morphology analysis of the tiny SN structure are very important to PD’s diagnosis and treatment evaluation. As the physiology of rhesus monkey is much similar to human’s, it has been used as an ideal experiment subject in many scientific researches.
OBJECTIVE: To modify traditional shape-based interpolation method for SN structure of the rhesus monkey brain slices.
METHODS: To gain the middle layers between two slices, traditional shape-based interpolation used linear weighted mean algorithm. However, considering the impact of adjacent multi-layer to the image, the middle layers were obtained by non-linear weight.
RESULTS AND CONCLUSION: The new interpolation algorithm was assessed with a favorable results. Also, this method was applied to the 3D reconstruction of SN structure. The results show that the new interpolation method can be used in nucleus group interpolation with sparse slices and tiny structures, and offer reference to the related disease research.