The hair cell analysis toolbox is a precise and fully automated pipeline for whole cochlea hair cell quantification

Abstract

Our sense of hearing is mediated by sensory hair cells, precisely arranged and highly specialized cells subdivided into outer hair cells (OHCs) and inner hair cells (IHCs). Light microscopy tools allow for imaging of auditory hair cells along the full length of the cochlea, often yielding more data than feasible to manually analyze. Currently, there are no widely applicable tools for fast, unsupervised, unbiased, and comprehensive image analysis of auditory hair cells that work well either with imaging datasets containing an entire cochlea or smaller sampled regions. Here, we present a highly accurate machine learning-based hair cell analysis toolbox (HCAT) for the comprehensive analysis of whole cochleae (or smaller regions of interest) across light microscopy imaging modalities and species. The HCAT is a software that automates common image analysis tasks such as counting hair cells, classifying them by subtype (IHCs versus OHCs), determining their best frequency based on their location along the cochlea, and generating cochleograms. These automated tools remove a considerable barrier in cochlear image analysis, allowing for faster, unbiased, and more comprehensive data analysis practices. Furthermore, HCAT can serve as a template for deep learning-based detection tasks in other types of biological tissue: With some training data, HCAT’s core codebase can be trained to develop a custom deep learning detection model for any object on an image.

PCP Auto Count: A Novel Fiji/ImageJ plug-in for automated quantification of planar cell polarity and cell counting

“…However, consistently obtaining high quality and similar SNR micrographs, especially across different samples, is often challenging. Indeed, this is a limitation inherent in several previously published cell quantification programs as well [26,37,38]. While we were able to successfully run PCPA on images of suboptimal quality (both in terms of image dpi and antibody SNR ratio) by using the pre-processing steps recommended, we have also included several features in the PCPA outputs (i.e.…”

“…A.I. approaches can also suffer from limited adaptability in the face of novel phenotypes or processing approaches not encountered during the training period [38]. Thus, while A.I.…”

“…Previously published automated cell counting programs also often struggle to cope with overlapping cells, which can require advanced machine learning systems [38], labor intensive optical clearing methods [45], or user guided segmentation of cells prior to machine analysis [26]. In the case of inner ear tissues this issue can be particularly problematic where, for example, there are little to no options for the labeling of cell nuclei in the neurons of the auditory nerve which makes segmentation nearly impossible.…”

Initiation and Formation of Stereocilia during the Development of Mouse Cochlear Hair Cells

“…Indeed, UN-RaPA can be used in combination with antibody staining to identify intersectional signatures of a protein with actin. A recent advance, HCAT, uses artificial intelligence and machine learning to identify and segment hair cells in the organ of Corti (Buswinka et al, 2023). By combining this with UN-RaPA to analyse the F-actin and other protein signatures, a powerful analytical pipeline can be developed for phenotyping across many samples, stained at different times in different labs.…”

“…Corti (Buswinka et al, 2023). By combining this with UN-RaPA to analyse the F-actin and other protein signatures, a powerful analytical pipeline can be developed for phenotyping across many samples, stained at different times in different labs.…”