A Novel Technique for Imaging and Analysis of Hair Cells in the Organ of Corti Using Modified Sca/eS and Machine Learning

Abstract

Here, we describe a sorbitol-based optical clearing method, called modified Sca/eS that can be used to image all hair cells (HCs) in the mouse cochlea. This modification of Sca/eS is defined by three steps: decalcification, de-lipidation, and refractive index matching, which can all be completed within 72 h. Furthermore, we established automated analysis programs that perform machine learning-based pattern recognition. These programs generate 1) a linearized image of HCs, 2) the coordinates of HCs, 3) a holocochleogram, and 4) clusters of HC loss. In summary, a novel approach that integrates modified Sca/eS and programs based on machine learning facilitates quantitative and comprehensive analysis of the physiological and pathological properties of all HCs.