Search bioRxiv⌕ Search

Biology subjects

Puria, S.

Publications and source records attributed to Puria, S..

10 recordsLinked to original sources

Sound-Driven Motion of Human Cochlear Structures Depends on Organ of Corti Integrity

Optical coherence tomography (OCT) was used to image and measure the passive sound-driven displacements of cochlear structures in situ through the intact round window membrane of fresh human cadaveric specimens at 12 to 30 hours postmortem. We find that the anatomical integrity of the organ of Corti (OoC) affects cochlear structural motion. OoC with high integrity (erect with supporting cells) have small osseous spiral lamina (OSL) motion. Conversely, OoC with low integrity (flattened without supporting cells) have large OSL motion. For the ratio of motion between the basilar membrane (BM) and OSL (BM/OSL), OoC with high integrity has large BM/OSL motion while a cochlear partition without OoC has small BM/OSL motion. Furthermore, the high-integrity OoC has reticular lamina surface motion that is lower than the BM, whereas low-integrity OoC has similar reticular lamina and BM motion. This relationship between OoC anatomical integrity and cochlear structural motion will have consequential effects on sound transduction.

neuroscience↗

A nonlinear time-domain finite-element model of the human middle ear subjected to intense sound waves

Computational models of the middle ear are typically developed within the linear range of the middle ear's response, which corresponds to typical sounds in daily life. Understanding middle-ear biomechanics under high-intensity sounds is also important for assessing potential auditory damage under adverse conditions. In this study, a finite-element (FE) model of the human middle ear plus the external ear canal is presented, employing a first-order Ogden hyperelastic material formulation and time-domain nonlinear analysis to capture nonlinear deformation of soft tissues and the resulting middle-ear response. The model was analyzed using pure-tone pressure excitations applied at the ear canal entrance over the 80 Hz to 8 kHz frequency range and 110 to 180 dB sound pressure levels (SPL). The ear canal pressure near the tympanic membrane (TM) and full-field three-dimensional response of the middle ear structures were examined. Across the investigated frequency range, the simulated middle-ear motion exhibited a transition from predominantly piston-like behavior at low frequencies, up to 1 kHz, to increasingly intricate multi-modal and multi-directional vibration patterns at higher frequencies. Nonlinear behaviors became increasingly significant at high frequencies and high stimulus levels, including compressive/expansive pressure-displacement relationships, waveform distortions, and the emergence of non-harmonic spectral components. The model developed in this study provides a computational framework for studying nonlinear sound transmission through the middle ear and its implications for high-intensity acoustic exposure. The findings contribute to a better understanding of middle-ear mechanics under adverse conditions and support future investigations of acoustic trauma and hearing protection strategies.

bioengineering↗

Sound from Ultrasound: Characterization of a MEMS-based personal-audio device in human temporal bones

Emerging MEMS-based audio devices generate ultrasonic acoustic output, but little is known about long-term biological effects of such exposures, which vary in frequency, intensity, duration, and coupling pathway. In this study, a novel transducer was characterized for its ultrasonic acoustic output, generating amplitude-modulated (AM) pressure pulses at a carrier frequency of approximately 200 kHz and operating frequency of approximately 100 kHz. To understand ultrasonic pressure wave transmission through the human middle and inner ear, sound pressure levels in the ear canal (PEC) and mechanical vibration velocities at the stapes (VST) and promontory (VPRM) were measured in human cadaveric temporal bones across frequencies up to 240 kHz. This represents the first such measurements at ultrasonic frequencies in human temporal bone specimens. Results demonstrated relatively consistent PEC measurements at ultrasonic frequencies across specimens, with minimal inter-specimen variation. At the operating frequency (~100 kHz), overall PEC was 82.2 dB SPL (SNR = 40.6 dB), while at the carrier frequency (~200 kHz), PEC increased to 90.6 dB SPL (SNR = 34.8 dB). VST and VPRM measurements at the operating frequency remained relatively independent of audio stimulus-drive frequency and voltage, with comparable magnitudes, suggesting that bone conduction pathways contribute significantly to inner ear ultrasonic exposure. When compared to existing safety guidelines, this device appears to meet specified criteria; however, given the limited scientific basis for these guidelines and unknown long-term effects, cautious application is recommended.

bioengineering↗

Anatomical integrity of the human cochlea estimated with optical coherence tomography for future clinical application

The human cochlea is encased within the otic capsule, the densest bone in the body, posing significant challenges for anatomical imaging of cochlear structures. Because of difficult access and fragility of cochlear structures, our understanding of intracochlear anatomy has historically relied on postmortem histology. We thus have a limited understanding of human cochlear anatomy in its native, unfixed state. Clinical diagnostics for hearing loss, such as audiometry and otoacoustic emissions, offer functional assessments but fail to elucidate the often diverse underlying structural pathologies with any degree of precision. To address the critical need for assessing the human cochlear anatomy and associated pathologies without the risk of traumatizing cochlear structures, we imaged fresh cochleae in situ soon after death through the intact round window membrane with Optical Coherence Tomography (OCT) without inserting instruments inside or opening the cochlea. Micron-resolution OCT cross-sectional images of the human intracochlear structures were acquired and compared with corresponding histology systematically to aid in the identification of fine structural features and possible pathologies. With OCT imaging, we observed varied anatomy of the organ of Corti, and developed a cochlear "integrity" rating system to differentiate healthy appearing cochleae from various pathological states. These results demonstrate the capability of OCT to non-traumatically visualize cochlear integrity, highlighting its potential as a diagnostic tool. This work shows promise in translating the ability to determine the likelihood of existing or lack of hair cells and supporting cells in live patients, which would enable appropriate targeted treatments.

physiology↗

From Simulations to Inference: Using Machine Learning to Tune Patient-Specific Finite-Element Models of the Middle Ear Towards Objective Diagnosis

Computational models, particularly finite-element (FE) models, are essential for interpreting experimental data and predicting system behavior, especially when direct measurements are limited. A major challenge in tuning these models is the large number of parameters involved. Traditional methods, such as one-by-one sensitivity analyses, are time-consuming, subjective, and often return only a single set of parameter values, focusing on reproducing averaged data rather than capturing the full variability of experimental measurements. In this study, we applied simulation-based inference (SBI) using neural posterior estimation (NPE) to tune an FE model of the human middle ear. The training dataset consisted of 10,000 FE simulations of stapes velocity, ear-canal (EC) input impedance, and absorbance, paired with seven FE parameter values randomly sampled within plausible ranges. The neural network learned the association between parameters and simulation outcomes, returning the probability distribution of parameter values that can reproduce experimental data. Our approach successfully identified parameter sets that reproduced three experimental datasets simultaneously. By accounting for experimental noise and variability during training, the method provided a probability distribution of parameters, representing all valid combinations that could fit the data, rather than tuning to averaged values. The network demonstrated robustness to noise and exhibited an efficient learning curve due to the large training dataset. SBI offers an objective alternative to laborious sensitivity analyses, providing probability distributions for each parameter and uncovering interactions between them. This method can be applied to any biological FE model, and we demonstrated its effectiveness using a middle-ear model. Importantly, it holds promise for objective differential diagnosis of conductive hearing loss by providing insight into the mechanical properties of the middle ear.

bioengineering↗

Computational model of the human cochlea with motion of the layered osseous spiral lamina

PurposeIn the base of the human cochlea, the partition anatomy is distinct from the commonly recognized anatomy of laboratory animals. The human features a radially wide, osseous spiral lamina (OSL) and a soft-tissue bridge region that connects the OSL to the basilar membrane proper. In addition to the basilar membrane, the human OSL and bridge move considerably. We investigated the complex cochlear partition in human emphasizing the layered structure of the OSL with a finite element model. Model results were calibrated with experimental measurements of motion from optical coherence tomography. MethodsThe box model contained two fluid chambers separated by a cochlear partition and a helicotrema. Model geometrical and material properties either came from literature, measurements, or were tuned to produce a frequency-place map for the passive human cochlea as well as motion results similar to experimental measurements. ResultsThe model motion results of the cochlear partition were similar to experimental results mostly within 5 dB but with differences at the high frequencies in both magnitude and phase beyond the best frequency. Around the best frequency location, the radial profile of cochlear partition motion was generally similar in both shape and magnitude. Sensitivity analysis, changing material-property parameters of the middle layer where the cochlear nerve fibers run between the layers of OSL plates, produced small changes in the model response and also showed negligible stress compared to the outer OSL plates. ConclusionThese results suggest that the layered OSL anatomy is favorable as a conduit and protection for the nerve fibers while simultaneously functioning as a mechanical lever.

biophysics↗

The impact of size on middle-ear sound transmission in elephants, the largest terrestrial mammal

Elephants have a unique auditory system that is larger than any other terrestrial mammal. To quantify the impact of larger middle ear (ME) structures, we measured 3D ossicular motion and ME sound transmission in cadaveric temporal bones from both African and Asian elephants in response to air-conducted (AC) tonal pressure stimuli presented in the ear canal (PEC). Results were compared to similar measurements in humans. Velocities of the umbo (VU) and stapes (VST) were measured using a 3D laser Doppler vibrometer in the 7-13,000 Hz frequency range, stapes velocity serving as a measure of energy entering the cochlea--a proxy for hearing sensitivity. Below the elephant ME resonance frequency of about 300 Hz, the magnitude of VU/PEC was an order of magnitude greater than in human, and the magnitude of VST/PEC was 5x greater. Phase of VST/PEC above ME resonance indicated that the group delay in elephant was approximately double that of human, which may be related to the unexpectedly high magnitudes at high frequencies. A boost in sound transmission across the incus long process and stapes near 9 kHz was also observed. We discuss factors that contribute to differences in sound transmission between these two large mammals.

physiology↗

A Cochlea-Slice Model using Floquet Boundary Conditions shows Global Tuning

A common assumption about the cochlea is that the local characteristic frequency (CF) is determined by a local resonance of basilar-membrane (BM) stiffness with the mass of the organ-of-Corti (OoC) and entrained fluid. We modeled the cochlea while avoiding such a priori assumptions by using a finite-element model of a 20-m-thick cross-sectional slice of the middle turn of a passive gerbil cochlea. The model had anatomically accurate structural details with physiologically appropriate material properties and interactions between the fluid spaces and solid OoC structures. The longitudinally-facing sides of the slice had a phase difference that mimicked the traveling-wave wavelength at the location of the slice by using Floquet boundary conditions. A paired volume-velocity drive was applied in the scalae at the top and bottom of the slice with the amplitudes adjusted to mimic experimental BM motion. The development of this computationally efficient model with detailed anatomical structures is a key innovation of this work. The resulting OoC motion was greatest in the transverse direction, stereocilia-tip deflections were greatest in the radial direction and longitudinal motion was small in OoC tissue but became large in the sulcus at high frequencies. If the source velocity and wavelength were held constant across frequency, the OoC motion was almost flat across frequency, i.e., the slice showed no local resonance. A model with the source velocity held constant and the wavelength varied realistically across frequency, produced a low-pass frequency response. These results indicate that tuning in the gerbil middle turn is not produced by a resonance due to local OoC mechanical properties, but rather is produced by the characteristics of the traveling wave, manifested in the driving pressure and wavelength. STATEMENT OF SIGNIFICANCEThe sensory epithelium of hearing, the organ of Corti, is encased in the bone of the fluid-filled cochlea and is difficult to study experimentally. We provide a new method to study the cochlea: making an anatomically-detailed finite-element model of a small transverse slice of the cochlea using Floquet boundary conditions and incorporating global cochlear properties in the slice drive and the wavelength-frequency relationship. The model shows that the slice properties do not show a mechanical resonance and therefore do not produce the frequency-response tuning of the cochlea. Instead, tuning emerges from global cochlear properties carried by the traveling wave.

biophysics↗

Motion of the Cochlear Reticular Lamina Varies Radially Across Outer-Hair-Cell Rows

The basilar membrane (BM) is connected to the reticular lamina (RL) through three rows of Y-shaped structures consisting of an outer hair cell (OHC) and a Deiters cell (DC) with a phalangeal process (PhP) that forms part of the RL mosaic surface. Morphological differences in the anatomy of the Y-shaped structures across the three OHC rows suggest differences in motion across the rows. Here we report OoC transverse motions measured across several radial locations for the gerbil basal region corresponding to ~45 kHz. Cross-sectional imaging and vibrometry measurements were made using a high-resolution (2.23 um axially in water) spectral-domain optical-coherence-tomography (SD-OCT) system. The stimuli were pure tones (2-63 kHz) at ear-canal sound pressure levels (SPLs) of 30-95 dB SPL in anesthetized gerbils (N=9) with healthy cochleae. We report displacements at the RL regions of OHC rows 1-3 (RL1-3), at the OHC-DC junctions of OHC rows 1-3 (OHC-DC-junction1-3), and at the arcuate zone, arcuate-pectinate junction, and pectinate zone of the BM (BMAZ, BMAPJ, and BMPZ, respectively). The in vivo BM displacements showed classic compressive nonlinearity and traveling-wave delays. The RL gain was similar to the BM gain at low frequencies (<20 kHz), but increased with frequency. Near the best frequency (BF), the RL gain was greater than the high-level BM gain by 40 {+/-}5 dB (mean{+/-}std), and had greater compressive nonlinearity. RL motion varied radially, and the RL3 gain was significantly greater than that of RL1 by 10 {+/-}1 dB (p<0.001). In contrast, the OHC-DC-junction gain varied little radially across OHCs. At low frequencies the OHC-DC-junction gain was constant across SPLs, and 14 {+/-} 3 dB greater than the BM gain. As the frequency increased, the OHC-DC-junction gain decreased to a level similar to the BM gain at BF. The RL2, 3 phase was advanced by 0.25-0.375 cycles relative to the BM phase at low frequencies, but the RL2, 3 phase lead decreased as the frequency increased, became similar to the BM phase at BF, and lagged behind the BM phase by 0.25-0.5 cycles above BF. The OHC-DC-junction phases were mostly similar to the BM phase at low frequencies, but became delayed relative to the BM as the frequency increased, typically by 0.25-0.5 cycles near BF and by up to 1 cycle above BF. Our results show the most detailed picture of motion around the three OHC rows yet published, indicating that RL motion varied radially. Surprisingly, there was little motion difference across the three OHC rows in the OHC-DC-junction region, indicating that the tops of the DCs move in unison. Our data show a rich array of OoC amplitude and phase variations that are not explained by current theories.

physiology↗

Watching Death in the Gerbil Cochlea Using Optical Coherence Tomography (OCT)

Because it is difficult to directly observe the morphology of the living cochlea, our ability to infer the mechanical functioning of the living ear has been limited. Nearly all of our knowledge about cochlear morphology comes from postmortem tissue that was fixed and processed using procedures that possibly distort the structures and fluid spaces of the organ of Corti. In this study, optical coherence tomography was employed to obtain in vivo and postmortem micron-scale volumetric images of the high-frequency hook region of the gerbil cochlea through the round-window membrane. The anatomical structures and fluid spaces of the organ of Corti were segmented and quantified in vivo and over a 90-minute postmortem period. The results show that some aspects of the organ of Corti are significantly altered over the course of death, such as the volumes of the fluid spaces, whereas the dimensions of other features change very little. We postulate that the fluid space of the outer tunnel and its surrounding tectal cells form a resonant structure that can affect the motion of the reticular lamina and thereby have a profound effect on outer-hair-cell transduction and thus cochlear amplification. In addition, the in vivo fluid pressure of the inner spiral sulcus is postulated to effectively inflate the connected sub-tectorial gap between the tectorial membrane and the reticular lamina. This gap height decreases after death, which is hypothesized to reduce and disrupt hair-cell transduction

neuroscience↗