3-D components of a biological neural network visualized in computer generated imagery. I. Macular receptive field organization

Computer-assisted, 3-dimensional reconstructions of macular receptive fields and of their linkages into a neural network have revealed new information about macular functional organization. Both type I and type II hair cells are included in the receptive fields. The fields are rounded, oblong, or elongated, but gradations between categories are common. Cell polarizations are divergent. Morphologically, each calyx of oblong and elongated fields appears to be an information processing site. Intrinsic modulation of information processing is extensive and varies with the kind of field. Each reconstructed field differs in detail from every other, suggesting that an element of randomness is introduced developmentally and contributes to endorgan adaptability.     

Lexical tone recognition with an artificial neural network

OBJECTIVES: Tone production is particularly important for communicating in tone languages such as Mandarin Chinese. In the present study, an artificial neural network was used to recognize tones produced by adult native speakers. The purposes of the study were (1) to test the sensitivity of the neural network to speaker variation typically in adult speaker groups, (2) to evaluate two normalization procedures to overcome the effects of speaker variation, and (3) to compare tone recognition performance of the neural network with that of the human listeners. DESIGN: A feedforward multilayer neural network was used. Twenty-nine adult native Mandarin Chinese speakers were recruited to record tone samples. The F0 contours of the vowel part of the 1044 monosyllabic words recorded were extracted using an autocorrelation method. Samples from the F0 contours were used as inputs to the neural network. The efficacy of the neural network was first tested by varying the number of inputs and the number of neurons in the hidden layer from 1 to 16. The sensitivity of the neural network to speaker variation was tested by (1) using the raw F0 data from speech tokens of a number of randomly drawn speakers that varied from 1 to 29, (2) using the raw F0 data from speech tokens of either male-only or female-only speakers, and (3) using two sets of normalized F0 data (i.e., tone 1-based normalization and first-order derivative) from speech tokens from a number of randomly drawn speakers that varied from 1 to 29. The recognition performance of the neural network under several experimental conditions was compared with the corresponding recognition performance of 10 normal-hearing, native Mandarin Chinese speaking adult listeners. RESULTS: Three inputs and four hidden neurons were found to be sufficient for the neural network to perform at about 85% correct using speech samples without normalization. The performance of the neural network was affected by variation across speakers particularly between genders. Using the tone 1-based normalization procedure, the performance of the neural network improved significantly. The recognition accuracy of the neural network as a whole or for each tone was comparable with that of the human listeners. CONCLUSIONS: The neural network can be used to evaluate the tone production of Mandarin Chinese speaking adults with human listener-like recognition accuracy. The tone 1-based normalization procedure improves the performance of the neural network to human listener-like accuracy. The success of our neural network in recognizing tones from multiple speakers supports its utility for evaluating tone production. Further testing of the neural network with hearing-impaired speakers might reveal its potential use for clinical evaluation of tone production.     

Isolated directional preponderance of caloric nystagmus: II. A neural network model

HYPOTHESIS: The purpose of this study was to simulate an isolated directional preponderance (DP) on bithermal caloric testing by constructing a realistic neural network model. The simulation was designed to capture not only the characteristics of the nystagmus response to caloric stimulation but also the response to rotational stimulation in patients with an isolated caloric DP. BACKGROUND: The nature of an isolated DP--that is, a DP in the absence of a significant spontaneous nystagmus or canal paresis--is outlined in the preceding article. In this article, the authors investigate the possible neural basis for an isolated caloric DP using the mathematic modeling technique of neural network simulation. Neural network models are typically abstract in nature; however, in this case the network was based on the known structure and function of the central vestibular system. METHODS: The neural network model was based on the known neuroanatomy and neurophysiology of the horizontal vestibuloocular reflex pathway. A leftward-rightward asymmetric modification of the dynamic responses of simulated medial vestibular nucleus type IA neurons on one side, or of type 2 neurons on the other side, to peripheral input would generate an isolated caloric DP. RESULTS: The values of DP and associated canal paresis produced by the network were within the same range as in the patient group. The network also predicted that the rotational DP would be lower than the caloric DP: between 2.5% and 56.9% of the caloric DP value. The actual rotational DP value was between 3% and 57% (average 41%) of the corresponding caloric DP value. CONCLUSIONS: An isolated caloric DP can be simulated by a neural network model by modifying the activity of model units that represent medial vestibular nucleus neurons. An asymmetric dynamic response by a gain-enhancement function of either type 1A neurons on one side or of type 2 neurons on the other was sufficient to produce an isolated caloric DP. Excitatory gain enhancement of type 2 neurons produced a smaller rotational DP than a similar modification of type 1 neurons. This result indicates a potential neural locus for the generation of an isolated DP in patients with vestibular disorders.     

Recognition of lexical tone production of children with an artificial neural network

Conclusion. This study demonstrated that the artificial neural network can successfully classify Mandarin Chinese tone patterns produced by multiple children. The neural network can be used as an objective way of evaluating tone production of children. Objectives. Traditionally, tone production is evaluated subjectively using human listeners. The aim of the present study was to investigate the efficacy of using an artificial neural network in evaluating tone production of Mandarin-speaking children. Subjects and methods. Speech materials were recorded from 61 normal-hearing children. The fundamental frequency (F0) of each monosyllabic word was extracted and then used as inputs to a feed-forward backpropagation artificial neural network. The number of inputs was set at 12, whereas the number of hidden neurons was set at 16 in the neural network. The output layer consisted of four neurons representing the four Mandarin tone patterns. The tone recognition performance of the neural network was further compared with that of native Mandarin-speaking adult listeners. Results. The neural network successfully classified the tone patterns of the 61 child speakers with an accuracy of about 85% correct. This high accuracy exceeded the tone recognition performance by the adult listeners. Individual child speakers showed varied tone production accuracy as recognized by the adult listeners or by the neural network.     

Recognition of lexical tone production of children with an artificial neural network

CONCLUSION: This study demonstrated that the artificial neural network can successfully classify Mandarin Chinese tone patterns produced by multiple children. The neural network can be used as an objective way of evaluating tone production of children. OBJECTIVES: Traditionally, tone production is evaluated subjectively using human listeners. The aim of the present study was to investigate the efficacy of using an artificial neural network in evaluating tone production of Mandarin-speaking children. SUBJECTS AND METHODS: Speech materials were recorded from 61 normal-hearing children. The fundamental frequency (F0) of each monosyllabic word was extracted and then used as inputs to a feed-forward backpropagation artificial neural network. The number of inputs was set at 12, whereas the number of hidden neurons was set at 16 in the neural network. The output layer consisted of four neurons representing the four Mandarin tone patterns. The tone recognition performance of the neural network was further compared with that of native Mandarin-speaking adult listeners. RESULTS: The neural network successfully classified the tone patterns of the 61 child speakers with an accuracy of about 85% correct. This high accuracy exceeded the tone recognition performance by the adult listeners. Individual child speakers showed varied tone production accuracy as recognized by the adult listeners or by the neural network.     

An artificial neural network improves prediction of observed survival in patients with laryngeal squamous carcinoma

The accepted method of modelling and predicting failure/survival, Cox’s proportional hazards model, is theoretically inferior to neural network derived models for analysing highly complex systems with large datasets. A blinded comparison of the neural network versus the Cox’s model in predicting survival utilising data from 873 treated patients with laryngeal cancer. These were divided randomly and equally into a training set and a study set and Cox’s and neural network models applied in turn. Data were then divided into seven sets of binary covariates and the analysis repeated. Overall survival was not significantly different on Kaplan–Meier plot, or with either test model. Although the network produced qualitatively similar results to Cox’s model it was significantly more sensitive to differences in survival curves for age and N stage. We propose that neural networks are capable of prediction in systems involving complex interactions between variables and non-linearity.     

Autofluorescence spectroscopy of oral leukoplakia using an artificial neural network

Objective. Early detection of oral cancer and premalignant lesions can improve the cure rate and reduce the morbidity of treatment. However, the incidence of oral cancer and precancer is low and risk groups are screened on a regular basis. Inspection is the most commonly used method of detection. Animal experiments have shown that premalignant lesions and squamous cell carcinoma of the oral mucosa can be detected by in‐vivo autofluorescence spectroscopy (AFS). Patients and methods. The experimental group consisted of 11 men and 10 women, mean age 60 years (range 30–81) In 22 oral leukoplakias AFS was performed, using a xenonlamp with a monochromator for illumination and an optical multichannel analyser for spectroscopy. The wavelength of the excitation light was 420 nm and AFS was recorded between 465 and 650 nm. Autofluorescence spectroscopy was measured in the centre and the border of the lesion, as well as in the healthy surrounding tissues and the contralateral counterpart. Data were analysed by using an artificial neural network. Results. The sensitivity of this system in the detection of oral leukoplakia was 86% with a specificity of 100%. With AFS homogeneous leukoplakia was recognized with a 73% sensitivity and 82% specificity. For non‐homogeneous lesions the sensitivity was 64% and the specificity 94%. In this material a correlation with histological classification has as yet not been found. Conclusion. This pilot study showed that oral leukoplakia can be detected with AFS. The artificial neural network was very valuable to analyse the data. Further research is justified to determine the value of AFS in the detection of oral cancer and premalignant lesions.     

Enhancement of acoustic evoked potentials and impairment of startle reflex induced by reduction of GABAergic control of the neural substrates of aversion in the inferior colliculus

The neural network of the inferior colliculus (IC), implicated in the generation of defensive behavior to aversive acoustic stimuli, is under tonic GABAergic control. Dopamine also seems to have a modulatory role in these neural circuits. It is still unclear how such changes in transmission of acoustic information influence the motor expression of the defensive behavior. Startle reaction to a sudden noise has been used as an effective way to measure the motor reactivity of rats to fearful acoustic stimuli. In this work we examined the processing of sensorial information--assessed by the recording of auditory evoked potentials (AEP)--and the behavioral effects--evaluated by the freezing and startle responses--during the reduction of GABA levels caused by microinjections of semicarbazide (SMC, 6 microg/0.2 microl), a glutamic acid decarboxylase inhibitor, into the IC. These data were compared to the effects of the overall arousal elicited by apomorphine (APO, 0.5 mg/kg, i.p.). The results obtained show that IC microinjections of SMC induced freezing behavior, enhanced the AEP and impaired the startle reaction to a loud sound. On the other hand, APO changed neither the AEP nor the startle in the same experimental conditions. These results suggest that the release of GABAergic control of the neural substrates of aversion in the IC results in an increased processing of auditory information along with an inhibitory influence on the motor pathways responsible for the startle response.     

Cortical neural activity underlying speech perception in postlingual adult cochlear implant recipients

Auditory event-related potentials (AERPs) and the simultaneously obtained behavioral measures (performance accuracy and reaction time) were used to study speech perception in postlingual adult cochlear implant (CI) recipients and in normal-hearing (NH) controls. AERPs were recorded while subjects were performing oddball discrimination tasks with increasing acoustic-phonetic demand. The tasks consisted of pairs of natural syllables that differed by one of the following phonetic contrasts: vowel place, voicing, vowel height, and place of articulation. Results indicated that the P3 potential was comparable in CI recipients and NH controls when the acoustic cues to the perception of the phonetic contrast were accessible. With the reduction in accessibility to the essential temporal and/or spectral cues, CI recipients exhibited delayed (prolonged P3 latency) and less synchronous (reduced amplitude) central speech-sound processing compared to NH controls. Among the phonetic contrasts used in the present study the place of articulation contrast yielded (1) the most prominent differences between CI recipients and NH controls across all measures, and (2) significant correlations between the neurophysiologic manifestation of speech discrimination (i.e. P3 latency), and conscious integration of perceptual information (i.e. performance accuracy and reaction time). Thus, P3 exposed the difficulties imposed on the impaired auditory system of CI recipients especially when elicited by speech contrasts that required processing of brief temporal-spectral cues. These findings support the P3 potential as a sensitive neural index of cortical processing that may provide information regarding accessibility and neural encoding of distinct acoustic-phonetic cues in CI recipients.     

Computer recognition of brain stem auditory evoked potential wave V by a neural network.

A neural network simulator was used for the recognition of the presence and location of the peak of wave V of the brain stem auditory evoked potential (BAEP) test. Waveforms selected from BAEPs performed in the last 4 years at the University of Pittsburgh Presbyterian University Hospital were digitized and sampled, and the resulting amplitudes were normalized. A training set was composed of the waveforms resulting from the stimulation of 50 ears. The normalized amplitudes were used as the initial activation values for the networks. The desired outputs (the target locations for wave V) were represented in the output layer by setting the output element, which corresponded to the target location and its immediate neighbors, to high activation levels, and all the remaining output units to zero activity. Two network architectures, differing only in the hidden unit layer, with 40 and 16 hidden units, respectively, were trained by using standard back-propagation. Several trials from different starting points were performed for each architecture. The best network, found after 60 epochs (3,000 presentations), was able to correctly identify 17 of 20 cases (85%) from a set of test cases that were independent from the training set.     

Response patterns and their relationship to frequency analysis in auditory forebrain centers of a songbird.

In the field L complex, the auditory part of the caudal telencephalon, multi-unit recordings were performed in seven awake, adult male starlings (Sturnus vulgaris, L). Pure tones in a frequency range between 0.5 and 6.0 kHz were used as stimuli. The field L complex of starlings consists of at least 11 functionally separated, tonotopically organized subcenters. The auditory processing of frequency information was investigated in eight of these areas. Two kinds of response patterns could be distinguished. The centers NA-L, NA2a, NA3 and NA4 showed phasic and sustained excitation, the other areas responded with phasic excitation only. All these auditory areas show strong tonotopic gradients, each of them representing the complete hearing range. The sharpest picture of the stimulus frequency is represented in the functional area NA-L. In relation to its total size, NA-L shows the smallest active area if stimulated by pure tones. In addition, in NA-L, only the excited neurons are surrounded by inhibited neurons during the response to a pure tone. This leads to an additional sharpening of frequency representation. In comparison with the other auditory areas, NA-L shows the greatest spatial extension of the tonotopic gradient. This, in combination with the smallest active size in NA-L, leads to the conclusion that in the primary projection field NA-L, the most neural space is available for the processing of a given frequency range.     

基于随机森林与人工神经网络慢性鼻窦炎伴鼻息肉诊断模型的构建与分析

目的 慢性鼻窦炎伴鼻息肉(CRSwNP)是一个全球性的健康问题,现有的诊断技术存在一定局限性,因此有必要开发新的诊断模型来补充现有的诊断方法。方法 利用CRSwNP患者(GSE23552、GSE36830)的公开基因表达数据来识别潜在的差异基因,应用随机森林算法和人工神经网络进一步筛选特异性基因,建立CRSwNP的早期诊断模型。结果 共发现78个上调基因和25个下调基因,随机森林算法筛选了4个特异性基因(HPGDS、IL1RL1、FMO3、DOK3),人工神经网络构建出基于上述基因的预测模型,该模型具有良好的预测效果(AUC=0.986),独立数据集GSE194282进一步验证了其准确性(AUC=0.888)。结论 采用机器学习方法建立了一个基于基因表达水平的预测模型,该模型可以预测早期CRSwNP,有助于早期诊断和改善临床决策。     

Morphological evidence for parallel processing of information in rat macula

Study of montages, tracings and reconstructions prepared from a series of 570 consecutive ultrathin sections shows that rat maculas are morphologically organized for parallel processing of linear acceleratory information. Type II cells of one terminal field distribute information to neighboring terminals as well. The findings are examined in light of physiological data which indicate that macular receptor fields have a preferred directional vector, and are interpreted by analogy to a computer technology known as an information network.     

Local haemodynamic changes associated with neural activity in auditory cortex

We used an optical technique to study haemodynamic changes associated with acoustically driven activity in auditory cortex of the chinchilla. Such changes are first detectable c. 0.5 s after stimulation, peak at 2-3 s, and decay within a further 3-6 s. This intrinsic signal imaging reveals activity in separate cortical areas, including primary auditory cortex (AI), secondary auditory cortex (AII) and an anterior auditory field (AAF). We have measured the timing of haemodynamics associated with each area, and find that AI has a different time course from AII and AAF; its haemodynamic change recovers more rapidly. We also show that within AI and AII, place specific activity related to acoustic stimulus frequency can be resolved by this optical imaging method. Our results show the close association between blood flow change and the local metabolic demands of neural activity. The data provide information about the potential of other functional imaging methods (e.g. PET, fMRI) which rely on activity related haemodynamic events.     

Auditory pathway encoding and neural plasticity in children with learning problems.

An inability to process auditory information, especially speech, characterizes many children with learning and attention problems. Our working hypothesis is that these speech-sound perception problems arise, at least in some cases, from faulty representation of the speech signal in central auditory centers. Preconscious neurophysiologic representation of sound structure by central auditory pathway neurons can be reflected by subcortical and cortical aggregate neural responses. These neurophysiologic responses can be modified by perceptual learning. Our research has shown that some children with learning problems demonstrate abnormal perception and neural representation of certain speech sounds. Differences between normal and learning-impaired groups can be attributable to aspects of neural synchrony that are reflected in aggregate neural responses. Deficiencies in neural synchrony in these children are apparent in subcortical (as well as cortical) representations of speech-sound structure, and these timing deficits are related to performance on speech-sound perception and learning measures. Moreover, impaired perception and neurophysiologic encoding of speech sounds can be improved with cue enhancement and can be modified by perceptual learning associated with auditory training.     

Evaluation of a model of the cochlear neural membrane. I. Physiological measurement of membrane characteristics in response to intrameatal electrical stimulation

To understand the auditory neural response to electrical stimuli similar to those used in a cochlear implant, it will be necessary to understand the neural refraction and summation response kinetics. Evidence exists indicating that the cell soma may alter the auditory neural response kinetics and could be the site of conduction failure for excitation initiated on the peripheral process. There is, however, reason to believe that the excitation site in some healthy, type I neurons and in pathological, type III neurons is the central process of the cell. To characterize the neural response to activation at a controlled central process site, cat auditory neurons were stimulated with an intrameatal electrode, and the summation and refraction response kinetics were measured. This approach was used to: (1) characterize the behavior of the neural response to central process excitation; (2) make comparisons between intrameatal excitation at a known central site and scala tympani excitation at an unknown site; and (3) provide membrane characterization free from the possible alteration of membrane kinetics produced by the cell soma. The membrane kinetics measured using intrameatal stimulation differ from those recorded with scala tympani stimulation indicating that the mechanisms for scala tympani and intrameatal stimulation differ.     

Auditory processing disorders: Relationship to cognitive processes and underlying auditory neural integrity

Background

Auditory processing disorder (APD) in children has been reported and discussed in the clinical and research literature for many years yet there remains poor agreement on diagnostic criteria, the relationship between APD and cognitive skills, and the importance of assessing underlying neural integrity.

Purpose

The present study used a repeated measures design to examine the relationship between a clinical APD diagnosis achieved with behavioral tests used in many clinics, cognitive abilities measured with standardized tests of intelligence, academic achievement, language, phonology, memory and attention and measures of auditory neural integrity as measured with acoustic reflex thresholds and auditory brainstem responses.

Method

Participants were 63 children, 7–17 years of age, who reported listening difficulties in spite of normal hearing thresholds. Parents/guardians completed surveys about the child's auditory and attention behavior while children completed an audiologic examination that included 5 behavioral tests of auditory processing ability. Standardized tests that examined intelligence, academic achievement, language, phonology, memory and attention, and objective tests auditory function included crossed and uncrossed acoustic reflex thresholds and auditory brainstem responses (ABR) were also administered to each child.

Results

Forty of the children received an APD diagnosis based on the 5 behavioral tests and 23 did not. The groups of children performed similarly on intelligence measures but the children with an APD diagnosis tended to perform more poorly on other cognitive measures. Auditory brainstem responses and acoustic reflex thresholds were often abnormal in both groups of children.

Summary

Results of this study suggest that a purely behavioral test battery may be insufficient to accurately identify all children with auditory processing disorders. Physiologic test measures, including acoustic reflex and auditory brainstem response tests, are important indicators of auditory function and may be the only indication of a problem. The results also suggest that performance on behavioral APD tests may be strongly influenced by the child's language levels.     

An infracuticular network is not required for outer hair cell shortening

The presence and distribution of filamentous actin was documented in normal and shortened isolated outer hair cells by use of fluorescently labeled phalloidin. In hair cells isolated from the guinea pig cochlea, an infracuticular network containing organized bundles of actin filaments is present in some, but not all, control cells and in some, but not all, cells after shortening. An infracuticular network of organized bundles of actin filaments is not present in control outer hair cells isolated from chinchilla or gerbil cochleas and it is never observed in any outer hair cells from these species even after shortening. Thus shortening of isolated outer hair cells is not correlated with the presence or the formation of organized bundles of actin filaments in an infracuticular network.     

珠海市新生儿听力普遍筛查网络数据库开发

目的利用网络数据库保存和管理新生儿听力筛查临床资料、筛查结果、听力学诊断和语言康复治疗情况,并探讨其应用价值。方法对我市全部"新生儿听力初次筛查不通过"资料进行系统管理,选用Access 2000程序建立"新生儿听力筛查Access 2000网络数据库"。数据库资料网络后台录入,用户名为新生儿家长手机号,提交后听力筛查、检测结果以短信方式通知新生儿家长,家长可以通过网络前台进行访问和就听力相关问题提问。结果网络数据库的建立简单、实用,运行情况良好。该数据库可快速完成全市新生儿听力筛查结果网络转诊、听力障碍新生儿跟踪随访、语言康复干预等功能。结论新生儿听力筛查Access 2000网络数据库对新生儿听力筛查资料具有强大的处理能力,便于各种数据的临床分析和疾病诊治,便于调查听力障碍新生儿流行病学分布特征和相关因素研究,便于网络医患相互交流。