Large-scale annotated dataset for cochlear hair cell detection and classification

Buswinka, C. J.; Rosenberg, D. B.; Simikyan, R. G.; Osgood, R. T.; Fernandez, K.; Nitta, H.; Hayashi, Y.; Liberman, L. W.; Nguyen, E.; Yildiz, E.; Kim, J.; Jarysta, A.; Renauld, J.; Wesson, E.; Thapa, P.; Bordiga, P.; McMurtry, N.; Llamas, J.; Kitcher, S. R.; Lopez-Porras, A. I.; Cui, R.; Behnammanesh, G.; Bird, J. E.; Ballesteros, A.; Velez-Ortega, A. C.; Edge, A. S.; Deans, M.; Gnedeva, K.; Shrestha, B. R.; Manor, U.; Zhao, B.; Ricci, A. J.; Tarchini, B.; Basch, M.; Stepanyan, R. S.; Landegger, L. D.; Rutherford, M.; Liberman, M. C.; Walters, B. J.; Kros, C.; Richardson, G. P.; Cunningham, L

doi:10.1101/2023.08.30.553559

Home/Archive/bioRxiv

bioRxiv · 10.1101/2023.08.30.553559

Large-scale annotated dataset for cochlear hair cell detection and classification

View original source ↗Export metadata ↓

Abstract

Our sense of hearing is mediated by cochlear hair cells, localized within the sensory epithelium called the organ of Corti. There are two types of hair cells in the cochlea, which are organized in one row of inner hair cells and three rows of outer hair cells. Each cochlea contains a few thousands of hair cells, and their survival is essential for our perception of sound because they are terminally differentiated and do not regenerate after insult. It is often desirable in hearing research to quantify the number of hair cells within cochlear samples, in both pathological conditions, and in response to treatment. However, the sheer number of cells along the cochlea makes manual quantification impractical. Machine learning can be used to overcome this challenge by automating the quantification process but requires a vast and diverse dataset for effective training. In this study, we present a large collection of annotated cochlear hair-cell datasets, labeled with commonly used hair-cell markers and imaged using various fluorescence microscopy techniques. The collection includes samples from mouse, human, pig and guinea pig cochlear tissue, from normal conditions and following in-vivo and in-vitro ototoxic drug application. The dataset includes over 90,000 hair cells, all of which have been manually identified and annotated as one of two cell types: inner hair cells and outer hair cells. This dataset is the result of a collaborative effort from multiple laboratories and has been carefully curated to represent a variety of imaging techniques. With suggested usage parameters and a well-described annotation procedure, this collection can facilitate the development of generalizable cochlear hair cell detection models or serve as a starting point for fine-tuning models for other analysis tasks. By providing this dataset, we aim to supply other groups within the hearing research community with the opportunity to develop their own tools with which to analyze cochlear imaging data more fully, accurately, and with greater ease.

Source connections

Published article ↗

Explore related subjects

neuroscience Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Buswinka, C. J., Rosenberg, D. B., Simikyan, R. G., Osgood, R. T., Fernandez, K., Nitta, H., Hayashi, Y., Liberman, L. W., Nguyen, E., Yildiz, E., Kim, J., Jarysta, A., Renauld, J., Wesson, E., Thapa, P., Bordiga, P., McMurtry, N., Llamas, J., Kitcher, S. R., Lopez-Porras, A. I., Cui, R., Behnammanesh, G., Bird, J. E., Ballesteros, A., Velez-Ortega, A. C., Edge, A. S., Deans, M., Gnedeva, K., Shrestha, B. R., Manor, U., Zhao, B., Ricci, A. J., Tarchini, B., Basch, M., Stepanyan, R. S., Landegger, L. D., Rutherford, M., Liberman, M. C., Walters, B. J., Kros, C., Richardson, G. P., Cunningham, L. 2023-09-01. Large-scale annotated dataset for cochlear hair cell detection and classification. https://doi.org/10.1101/2023.08.30.553559

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

Distinguishing evidence accumulation from choice confidence in neural signatures of perceptual decision making

Human electrophysiology research has identified the centro-parietal positivity (CPP) as an apparent neural correlate of sensory evidence accumulation during perceptual decision-making. The CPP has also been shown to closely scale with subjective confidence. Coupled with the fact of its domain-generality and lack of choice selectivity, this observation raises the possibility that the CPP might in fact trace subjective confidence rather than cumulative evidence, two quantities that are closely correlated in most experiments. To disentangle confidence from cumulative evidence, we used predictive cues with 80% validity to manipulate prior expectations of participants during a contrast discrimination task while recording electroencephalography (EEG). At the time of decision termination, confidence is expected to reach a higher level for validly cued versus invalidly cued choices, whereas the opposite would be expected of a cumulative evidence signal due to biased decision starting points. As expected, valid cues led to faster, more accurate and more confident responses compared to invalid and neutral cues. At the neural level, cues produced systematic increases in preparation of the expected action prior to evidence onset as indexed by premotor mu/beta activity, consistent with cue-biased criterion adjustments. The pre-response CPP amplitude was significantly reduced for validly cued choices and the effect remained even when matching for confidence. Together these findings suggest that the CPP is not a direct read-out of subjective confidence, but an evidence accumulation signal that contributes to both decision formation and confidence estimation.

neuroscience↗

bioRxiv2026

A Neurocompliant Ionic Conductive Hydrogel Orchestrates Glial and Vascular Remodeling to Mitigate Traumatic Brain Injury

Developing conductive scaffolds for neural regeneration is often hindered by the cytotoxicity and mechanical mismatch inherent to conventional electronic fillers. Here, we report an ultrasoft ionically conductive hydrogel engineered by grafting choline-based bioionic liquids (BIL) onto fish gelatin (FiBGel). Systematic optimization yields brain-mimetic conductivity (0.25-0.68 S/m), elastic moduli (0.6-0.9 kPa), and physiological water content (72-81%). In 3D culture, the matrix drives spontaneous neurogenesis, achieving ~90% MAP2+ neurons with robust calcium signaling, while concurrently supporting endothelial network formation. We report the first intracerebral administration of a BIL hydrogel, demonstrating remarkable biocompatibility and immunomodulation. In a traumatic brain injury model, a single FiBGel dose halves reactive astrogliosis, enhances cortical revascularization, and limits lesion cavity expansion. Leveraging an FDA-recognized micronutrient, FiBGel offers a translatable strategy that integrates neuromechanics, conductivity, immune modulation, and vascular support within one injectable scaffold for minimally invasive treatment of brain injuries.

neuroscience↗

bioRxiv2026

Changes in anterior cingulate connectivity associated with response to antidepressant treatment for major depressive disorder

Background: Resting-state functional connectivity (rsFC) involving the anterior cingulate cortex (ACC) and the hippocampal formation has been implicated in major depressive disorder (MDD) and antidepressant response. However, it remains unclear whether connectivity changes in distinct ACC subdivisions and hippocampal formation reflect general antidepressant effects or are specifically associated with clinical improvement. Methods: Forty-one, unmedicated adults with MDD underwent functional MRI before and after 8 weeks of sertraline or bupropion treatment. Seed-based rsFC analyses examined subgenual ACC (sgACC), pregenual ACC (pgACC), dorsal ACC (dACC), and hippocampal formation connectivity. Linear mixed-effects models tested whether connectivity change was associated with treatment response, remission, and dimensional improvement in depressive symptoms. Results: Thirty individuals (73.2%) responded to treatment and 26 (63.4%) achieved remission. Overall, rsFC decreased pre- to post-treatment between (1) dACC and pgACC (but not sgACC and hippocampal formation) and (2) cortical regions primarily involving default mode and frontoparietal networks. When considering treatment response, differences in sgACC and pgACC connectivity with temporal and inferior frontal regions emerged (increased in responders vs. decreased in non-responders). Remission was similarly associated with increased sgACC connectivity with temporal and prefrontal regions, and increased sgACC connectivity with temporal cortical regions related to greater reductions in depressive symptoms. Conclusions: Antidepressant treatment was associated with distinct patterns of ACC connectivity change, with general reductions in dorsal and pregenual connectivity but increases in subgenual and pregenual connectivity associated with favorable clinical outcomes. Findings suggest that changes in sgACC/pgACC connectivity are particularly relevant to antidepressant treatment response. Keywords: Major depressive disorder; Resting-state functional connectivity; Anterior cingulate cortex; Antidepressant treatment; Treatment response; Subgenual anterior cingulate cortex.

neuroscience↗