An image segmentation method based on the spatial correlation coefficient of Local Moran’s I - identification of A-type potassium channel clusters in the thalamus

• DOI: 10.7554/elife.89361.2
• Publication Date: 2024-09-13T13:26:00Z
• Authors: Csaba Dávid; Kristóf Giber; Katalin Kerti-Szi...; Mihaly Kollo; Zoltán Nusser; László Acsády
• Keywords: Spatial correlation

Unsupervised segmentation in biological and non-biological images is only partially resolved. Segmentation either requires arbitrary thresholds or large teaching datasets. Here we propose a spatial autocorrelation method based on Local Moran’s I coefficient to differentiate signal, background noise any type of image. The method, originally described for geoinformatics, does not require predefined intensity threshold algorithm image allows quantitative comparison samples obtained different conditions. It utilizes relative as well information neighboring elements select spatially contiguous groups pixels. We demonstrate that outperforms threshold-based (TBM) both artificially generated natural especially when substantial. This superior performance can be attributed the exclusion false positive pixels resulting from isolated, high To test method’s power real situation used confocal somatosensory thalamus immunostained Kv4.2 Kv4.3 (A-type) voltage gated potassium channels. identified ion channel clusters thalamic neuropil. Spatial distribution these displayed strong correlation with sensory axon terminals subcortical origin. unique association special presynaptic postsynaptic cluster was confirmed electron microscopy. These data rapid, simple optimal variable nose