# VASCilia is an open-source, deep learning-based tool for 3D analysis of cochlear hair cell stereocilia bundles

## TL;DR

VASCilia is an open-source tool that uses deep learning to automate 3D analysis of cochlear hair cell structures, aiding hearing research.

## Contribution

VASCilia introduces a deep learning-based, open-source tool for high-throughput 3D analysis of cochlear hair cell stereocilia bundles.

## Key findings

- VASCilia includes five deep learning models for tasks like Z-focus tracking, alignment, segmentation, position prediction, and classification.
- The tool successfully quantified bundle disorganization in Cdh23−/− cochleae, showing more heterogeneous and irregular bundles compared to controls.
- A manually annotated dataset of 55 3D stacks with 2,205 hair cell bundles is provided as a foundational resource for future research.

## Abstract

Cochlear hair cells are essential for hearing, and their stereocilia bundles are critical for mechanotransduction. However, analyzing the 3D morphology of these bundles can be challenging due to their complex organization and the presence of other cellular structures in the tissue. To address this, we developed VASCilia (Vision Analysis StereoCilia), a Napari plugin suite that automates the analysis of 3D confocal microscopy datasets of phalloidin-stained cochlear hair cell bundles. VASCilia includes five deep learning-based models trained on mouse cochlear datasets that streamline the analysis process, including: (1) Z-Focus Tracker (ZFT) for selecting relevant slices in a 3D image stack; (2) PCPAlignNet (Planar Cell Polarity Alignment Network) for automated orientation of image stacks; (3) a segmentation model for identifying and delineating stereocilia bundles; (4) a tonotopic Position Prediction tool; and (5) a classification tool for identifying hair cell subtypes. In addition, VASCilia provides automated computational tools and measurement capabilities. Using VASCilia, we demonstrate its utility on challenging datasets, including neonatal wild type and Eps8 KO 5-day old mice. We further showcase its power by quantifying complex bundle disorganization in Cdh23−/− cochleae via texture analysis, which revealed systematically more heterogeneous and less regular bundles than littermate controls. These case studies demonstrate the power of VASCilia in facilitating detailed quantitative analysis of stereocilia. VASCilia also provides a user-friendly interface that allows researchers to easily navigate and use the tool, with the added capability to reload all their analyses for review or sharing purposes. We believe that VASCilia will be a valuable resource for researchers studying cochlear hair cell development and function, addressing a longstanding need in the hair cell research community for specialized deep learning-based tools capable of high-throughput image quantitation. We have released our code along with a manually annotated dataset that includes approximately 55 3D stacks featuring instance segmentation (https://github.com/ucsdmanorlab/Napari-VASCilia). This dataset comprises a total of 502 inner and 1,703 outer hair cell bundles annotated in 3D. As the first open-source dataset of its kind, we aim to establish a foundational resource for constructing a comprehensive atlas of cochlea hair cell images. Ultimately, this initiative will support the development of foundational models adaptable to various species, markers, and imaging scales to accelerate advances within the hearing research community.

Cochlear hair cell stereocilia bundles are vital for hearing, but their 3D morphology remains unclear due to their complex organisation. This study develops an open-source, deep learning-based tool called VASCilia that automates the high-throughput analysis of 3D confocal microscopy datasets of phalloidin-stained cochlear hair cell bundles.