检索结果-内蒙古大学图书馆

33rd International Florida Artificial Intelligence Research Society Conference, FLAIRS 2020

作者： Wisler, Alan Teplansky, Kristin Green, Jordan R. Austin, Sara G. Wang, Jun Department of Communication Sciences and Disorders University of Texas at Austin AustinTX United States Department of Communication Sciences and Disorders Mgh Institute of Health Professions BostonMA United States Speech and Hearing Bioscience and Technology Program Harvard University BostonMA United States Department of Neurology Dell Medical School University of Texas at Austin AustinTX United States

ISBN: (纸本)9781577358213

In this study we examined the efficacy of machine learning general regression algorithms for predicting ordinal variables based on the acoustic speech signal. We were specifically interested in whether predictions that fell between ordinal levels (e.g. a predicted score of 3.2 instead of a true score of 3) contained meaningful information about the outcome variable. As a test case, we explored speech-based estimation of the Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R), a clinical measure for speech severity. The ALSFRS-R is a diagnostic tool that measures individual components of motor function for patients with ALS along a 5-point ordinal scale. Using artificial neural networks (ANN) we can generate continuous estimates of the speech component in ALSFRS-R. However, the degree to which the improved resolution of these estimates contains useful information related to patients' motor control has not been thoroughly studied. In this paper, we sought to answer this question by comparing the residuals in machine learning estimates of the ALSFRS-R speech score with patients' intelligible speaking rate (ISR), which is a more granular measure of speech motor control. Experimental results using speech data from 45 patients with ALS confirmed that ANN regression is effective at learning granular information even when trained with coarse ordinal labels. © FLAIRS *** right reserved.

关键词： speech

来源：评论

学校读者我要写书评

暂无评论

Characterizing Vocal Hyperfunction Using Ecological Momentary Assessment of Relative Fundamental Frequency

引用

Journal of Voice 2024年

作者： Cheema, Ahsan J. Marks, Katherine L. Ghasemzadeh, Hamzeh Van Stan, Jarrad H. Hillman, Robert E. Mehta, Daryush D. Speech and Hearing Bioscience and Technology Program Harvard University 25 Shattuck St Boston 02115 Massachusetts United States Harvard Medical School 25 Shattuck St Boston 02115 Massachusetts United States Mass General Hospital (MGH) Voice Center 1 Bowdoin Sq Boston 02114 Massachusetts United States Eaton-Peabody Laboratories Massachusetts Eye and Ear 243 Charles St Boston 02114 Massachusetts United States Sargent College of Health and Rehabilitation Sciences Boston University 677 Beacon St Boston 02215 Massachusetts United States MGH Institute of Health Professions 36 1st Ave Boston 02129 Massachusetts United States

Many common voice disorders are associated with vocal hyperfunction (VH), with subtypes including phonotraumatic VH (leading to organic vocal fold lesions such as nodules and/or polyps) and nonphonotraumatic VH (often diagnosed as primary muscle tension dysphonia). VH has been hypothesized to influence baseline vocal fold tension during phonation, and the relative fundamental frequency (RFF) during onset and offset cycles of phonation has been related to vocal fold tension and has been shown to differentiate typical voices from patients with VH in laboratory settings. In this study, we investigated whether the laboratory sensitivity of RFF to the presence of VH found in the laboratory is preserved in naturalistic, in-field settings and whether ecological momentary assessment of RFF during daily life could be a correlate of self-reported vocal effort. RFF analysis was carried out after performing smartphone-based monitoring of anterior neck-surface vibration with accelerometer sensors in both laboratory and in-field settings. Supervised machine learning was applied to combine multiple RFF values to discriminate and classify patients with VH from vocally typical speakers. Results showed that RFF-based classification of VH can be preserved in the naturalistic environments for patients with phonotraumatic (81.3% accuracy) and nonphonotraumatic (62.5% accuracy) VH. Additionally, we used explainability techniques to understand which RFF features were clinically relevant in the classification tasks. No direct relationship was observed between RFF and self-reported vocal effort. Overall, this study advances our understanding about RFF as a potential biomarker of VH as individuals go about their daily life. Machine learning algorithms can be implemented within a monitoring device for proactive screening or in biofeedback-based voice therapy paradigms. © 2024 The Authors

关键词： Relative fundamental frequency—Vocal hyperfunction—Phonotraumatic vocal hyperfunction—Nonphonotraumatic vocal hyperfunction—Vocal effort—Machine learning—Random forest

来源：评论

学校读者我要写书评

暂无评论

fMRI reveals language-specific predictive coding during naturalistic sentence comprehension

引用

Neuropsychologia 2020年第0期138卷 107307-000页

作者： Shain, Cory Blank, Idan Asher van Schijndel, Marten Schuler, William Fedorenko, Evelina The Ohio State University 43210 United States University of California Los Angeles 90024 United States Massachusetts Institute of Technology 02139 United States Cornell University 14853 United States Massachusetts General Hospital Program in Speech and Hearing Bioscience and Technology 02115 United States

Much research in cognitive neuroscience supports prediction as a canonical computation of cognition across domains. Is such predictive coding implemented by feedback from higher-order domain-general circuits, or is it locally implemented in domain-specific circuits? What information sources are used to generate these predictions? This study addresses these two questions in the context of language processing. We present fMRI evidence from a naturalistic comprehension paradigm (1) that predictive coding in the brain's response to language is domain-specific, and (2) that these predictions are sensitive both to local word co-occurrence patterns and to hierarchical structure. Using a recently developed continuous-time deconvolutional regression technique that supports data-driven hemodynamic response function discovery from continuous BOLD signal fluctuations in response to naturalistic stimuli, we found effects of prediction measures in the language network but not in the domain-general multiple-demand network, which supports executive control processes and has been previously implicated in language comprehension. Moreover, within the language network, surface-level and structural prediction effects were separable. The predictability effects in the language network were substantial, with the model capturing over 37% of explainable variance on held-out data. These findings indicate that human sentence processing mechanisms generate predictions about upcoming words using cognitive processes that are sensitive to hierarchical structure and specialized for language processing, rather than via feedback from high-level executive control mechanisms. © 2019 Elsevier Ltd

关键词： fMRI Language Multiple demand network Naturalistic Predictive coding Sentence processing Surprisal Syntactic structure

来源：评论

学校读者我要写书评

暂无评论

Higher-order comparisons of sentence encoder representations

arXiv

引用

arXiv 2019年

作者： Abdou, Mostafa Kulmizev, Artur Hill, Felix Low, Daniel M. Søgaard, Anders Department of Computer Science University of Copenhagen Department of Linguistics and Philology Uppsala University DeepMind Program in Speech and Hearing Bioscience and Technology Harvard Medical School-MIT

Representational Similarity Analysis (RSA) is a technique developed by neuroscientists for comparing activity patterns of different measurement modalities (e.g., fMRI, electrophysiology, behavior). As a framework, RSA has several advantages over existing approaches to interpretation of language encoders based on probing or diagnostic classification: namely, it does not require large training samples, is not prone to overfitting, and it enables a more transparent comparison between the representational geometries of different models and modalities. We demonstrate the utility of RSA by establishing a previously unknown correspondence between widely-employed pretrained language encoders and human processing difficulty via eye-tracking data, showcasing its potential in the interpretability toolbox for neural models. Copyright © 2019, The Authors. All rights reserved.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

Preface: To the Ear and Back Again-Advances in Auditory Biophysics (Proceedings of the 13th Mechanics of hearing Workshop)

引用

AIP Conference Proceedings 2018年第1期1965卷

作者： Puria, Sunil Department of Otolaryngology Harvard Med. School Harvard University Graduate School of Arts and Sciences Director of Admissions Harvard MIT Program in Speech and Hearing Bioscience and Technology -Lit ENGL United States

Preface: To the Ear and Back Again - Advances in Auditory Biophysics (Proceedings of the 13th Mechanics of hearing Workshop), AIP Conference Proceedings, Volume

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Role of Tumor Necrosis Factor Alpha (TNFα) in hearing Loss and Vestibular Schwannomas

引用

Current Otorhinolaryngology Reports 2018年第1期6卷 15-23页

作者： Ren, Yin Stankovic, Konstantina M. Department of Otolaryngology Massachusetts Eye and Ear 243 Charles Street Boston 02114 MA United States Department of Otolaryngology Harvard Medical School 25 Shattuck St Boston 02115 MA United States Eaton Peabody Laboratories Massachusetts Eye and Ear 243 Charles Street Boston 02114 MA United States Harvard Program in Speech and Hearing Bioscience and Technology 25 Shattuck Street Boston 02115 MA United States

Purpose of Review: The aim of this review is to highlight relevant literature on the role of tumor necrosis factor alpha (TNFα) in sensorineural hearing loss (SNHL) and vestibular schwannomas (VS). Recent Findings: A comprehensive review of publically available databases including PubMed was performed. The mechanism by which hearing loss occurs in VS is still unknown and likely multifactorial. Genetic differences between VSs and tumor-secreted proteins may be responsible, at least in part, for VS-associated SNHL. TNFα has pleotropic roles in promoting inflammation, maintaining cellular homeostasis, inducing apoptosis, and mediating ototoxicity in patients with sporadic VS. TNFα-targeted therapies have shown efficacy in both animal models of sensorineural hearing loss and clinical trials in patients with immune-mediated hearing loss. Efforts are underway to develop nanotechnology-based methods to target TNFα and other pathogenic molecules in VS. Summary: Development of molecularly targeted therapies against TNFα represents an important area of research in ameliorating VS-associated hearing loss. © 2018, Springer Science+Business Media, LLC, part of Springer Nature.

关键词： hearing loss Inflammation Nanotechnology TNF alpha TNFα vestibular schwannoma

来源：评论

学校读者我要写书评

暂无评论

Restoration of visual function in adult mice with an inherited retinal disease via adenine base editing ()

引用

NATURE BIOMEDICAL ENGINEERING 2020年第11期4卷 1119-1119页

作者： Suh, Susie Choi, Elliot H. Leinonen, Henri Foik, Andrzej T. Newby, Gregory A. Yeh, Wei-Hsi Dong, Zhiqian Kiser, Philip D. Lyon, David C. Liu, David R. Palczewski, Krzysztof Gavin Herbert Eye Institute Department of Ophthalmology University of California Irvine Irvine CA USA Department of Pharmacology Case Western Reserve University Cleveland OH USA Department of Physiology and Biophysics University of California Irvine Irvine CA USA Department of Chemistry University of California Irvine Irvine CA USA Department of Anatomy and Neurobiology School of Medicine University of California Irvine Irvine CA USA International Centre for Translational Eye Research Institute of Physical Chemistry Polish Academy of Sciences Warsaw Poland Merkin Institute of Transformative Technologies in Healthcare Broad Institute of Harvard and MIT Cambridge MA USA Department of Chemistry and Chemical Biology Harvard University Cambridge MA USA Howard Hughes Medical Institute Harvard University Cambridge MA USA Program in Speech and Hearing Bioscience and Technology Harvard Medical School Boston MA USA Research Service VA Long Beach Healthcare System Long Beach CA USA

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Impedance measurements of the human cochlear partition

引用

AIP Conference Proceedings 2018年第1期1965卷

作者： Stefan Raufer Hideko H. Nakajima 1Program in Speech and Hearing Bioscience and Technology Harvard University USA 2Dept. of Otolaryngology Harvard Medical School Massachusetts Eye and Ear Boston MA USA

The cochlea is a mechanical frequency analyzer, owing its characteristics to the impedance of the cochlear partition. In humans, the impedance of the partition has not been measured directly, and estimates of the stiffness (a principal component of the impedance) are based on loose assumptions. In this study, we examine not only the stiffness of the basilar membrane (BM), but also the osseous spiral lamina (OSL), which, in human, vibrates substantially. We hypothesize that the OSL contributes significantly to the volume stiffness of the cochlear partition (CP). We measured velocities of the BM and OSL at different radial locations 1 mm from the base of the cochlea in a fresh human cadaveric specimen. Simultaneously, we measured intracochlear pressures on the other side of the partition, in scala vestibuli. With the velocity and pressure measurements we can estimate the specific acoustic impedance of the BM and OSL (Z = p/v). At frequencies well below the resonant frequency, the stiffness of these structures can be extracted by multiplying the impedance by the radian frequency. The specific acoustic stiffness was found to be 1.2 GPa/m on the BM, 6 GPa/m at the juncture where the BM attaches to the OSL, and 10 GPa/m at the midpoint of the OSL. A beam model, appropriate to model the radial motion of the BM in guinea pig or gerbil, cannot describe the displacement of the human CP in the base. Instead, we find that the OSL is hinged near the modiolus and vibrates significantly near the connection to the more compliant BM, contributing greatly the volume compliance of the CP.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Stimulus frequency otoacoustic emission generation within a finite element model of the mouse cochlea: The effect of impedance irregularities of organ of Corti structures

引用

AIP Conference Proceedings 2018年第1期1965卷

作者： Hamid Motallebzadeh Sunil Puria 1Eaton-Peabody Laboratory Massachusetts Eye and Ear 2Department of Otolaryngology Harvard Medical School 3Speech and Hearing Bioscience and Technology Harvard University Graduate School of Arts and Sciences

Stimulus-Frequency Otoacoustic emissions (SFOAEs) are sounds that are generated by the cochlea and carry signatures of cochlear mechanisms and amplification. Although SFOAEs are not thought to contribute to the sensitivity of the hearing process, they provide valuable non-invasive information about cochlear function and potentially can be used in the clinical assessment of sensory-neural hearing loss. The source of SFOAE is thought to be due to the travelling wave reflected by local impedance roughness which is due to random perturbations of the organ of Corti structures (Shera and Zweig, 1993). While the active process of the outer hair cells (OHCs) tunes the peak of the traveling wave, the source of the random perturbations in generating SFOAEs is not well established. In the present work, random perturbations around baseline model parameters (e.g., Young’s moduli and geometrical parameters of the basilar membrane (BM), reticular lamina (RL), outer hair cells (OHCs) as well as their gain factor (the ratio of the axial force that OHC generates due to the shear force on OHC stereocilia) were studied in a finite-element model of the mouse cochlea. Consistent with experimental measurements, the model successfully produces SFOAE magnitudes and phase delays, without significantly affecting cochlear tuning. The model SFOAE amplitude and phase are more sensitive to perturbations in BM Young’s modulus and thickness and to OHC gain than to RL Young’s modulus and thickness and OHC Young’s moduli. Irregularities of OHC gain factor seems to be a plausible source of the SFOAEs because not only their distribution varies along the cochlear length but also their functionality is vulnerable. The model also confirms that the SFOAE magnitude increases, approximately linearly with the perturbation of BM Young’s modulus.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Drive mechanisms to the inner and outer hair cell stereocilia

引用

AIP Conference Proceedings 2018年第1期1965卷

作者： Nima Maftoon Hamid Motallebzadeh John J. Guinan, Jr. Sunil Puria 1Eaton-Peabody Laboratory Massachusetts Eye and Ear Harvard University Graduate School of Arts and Sciences USA 2Department of Otolaryngology Harvard Medical School Harvard University Graduate School of Arts and Sciences USA 3Speech and Hearing Bioscience and Technology Harvard University Graduate School of Arts and Sciences USA

It has been long believed that inner hair cell (IHC) stimulation can be gleaned from the classic ter-Kuile shear motion between the reticular lamina (RL) and tectorial membrane (TM). The present study explores this and other IHC stimulation mechanisms using a finite-element-model representation of an organ of Corti (OoC) cross section with fluid-structure interaction. A 3-D model of a cross section of the OoC including soft tissue and the fluid in the sub-tectorial space, tunnel of Corti and above the TM was formulated based on anatomical measurements from the gerbil apical turn. The outer hair cells (OHCs), Deiter’s cells and their phalangeal processes are represented as Y-shaped building-block elements. Each of the IHC and OHC bundles is represented by a single sterocilium. Linearized Navier-Stokes equations coupled with linear-elastic equations discretized with tetrahedral elements are solved in the frequency domain. We evaluated the dynamic changes in the OoC motion including sub-tectorial gap dimensions for 0.1 to 10 kHz input frequencies. Calculations show the classic ter-Kuile motion but more importantly they show that the gap-height changes which produce oscillatory radial flow in the subtectorial space. Phase changes in the stereocilia across OHC rows and the IHC are also observed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：