Listing 1 - 10 of 16 | << page >> |
Sort by
|
Choose an application
Choose an application
Frequency selectivity and neural phase-locking are two primary properties of the peripheral auditory system and are fundamental to hearing. These properties are reflected in neural signals as a place-rate and time code and are quantified by means of responses from single auditory nerve fibers (AN fibers). Because of practical issues, the limits of both properties are poorly characterized in humans and widely differing values are assumed in the interpretation of non-electrophysiological results. Recent otoacoustic and behavioral measurements suggest that frequency selectivity is higher in humans than in laboratory animals (Shera et al., 2002), but this is disputed based on comparisons of behavioral and electrophysiological measurements across species (Ruggero and Temchin, 2005). Neural phase-locking declines with frequency and becomes undetectable at an upper frequency limit which differs between species. Some studies suggest that phase-locking in human extends to much higher frequencies than that found in common laboratory animals such as cat. In this thesis we investigate the use of mass potentials to electrophysiologically quantify these properties in humans.A minimally invasive transtympanic protocol was developed to record stimulus-evoked mass-potentials from the cochlear promontory or from the niche of the round window in monkey (macaque) and human. This involved a custom made ear mold with openings for the transtympanic needle electrode and for acoustic stimulation (ER-2 earphone, in situ calibrated with an ER-7 microphone). To obtain frequency selectivity a notched-noise forward masking paradigm (Oxenham and Shera, 2003) was combined with the recording of compound action potentials (CAP). For neural phase-locking, we developed a method based on forward masking to disambiguate phase-locked contributions of receptor and neural origin (neurophonic). Both methods were first developed, assessed and validated at the round window in cat and thereafter applied in monkey and human.Frequency selectivity. We found that the empirical method that was first developed and assessed in cat was a suitable means to examine frequency tuning for a number of reasons. a) Masking of CAP responses behaved as expected from the physiology of single AN-fibers. b) Masking tuning curves were equivalent to tuning curve of single AN-fibers. c) The sharpness of masking tuning curves, quantified as a Q10-factor, closely followed the lower boundary of single AN-fibers and was only slightly higher at high frequencies. CAP-tuning was dependent on probe level consistent with cochlear physiology. The CAP responses behaved similar in human and monkey, but the responses and signal-to-noise ratios were much smaller and therefore higher stimulus levels (>30 dB) were needed in human and monkey. We found that CAP tuning in human was on average 1.6 times sharper than in cat and chinchilla and 1.35 times sharper than in monkey (macaque).Neural phase-locking. We found that potentials near the round window have a considerable neural phase-lockedcontribution that can be isolated and quantified with the present method. Moreover, the neural component was in many ways similar to that measured on the auditory nerve. The only noteworthy differences with the latter were a typically larger magnitude and a smaller time lag for the neural component at the round window. The frequency limit obtained in cat was 4.7 kHz, which is very similar to the limit reported for individual AN-fibers (~ 5 kHz). We found in many respects similarities between the measurements in human, monkey and those at the round window in cat, with similar temporal and spectral properties. The most notable difference in human and monkey compared to cat was a much smaller magnitude (-30 dB) and also signal-to-noise(floor) ratio (-20 dB) for the neurophonic. The upper detectable frequency limits of neural phase-locking obtained from stimulus evoked potentials was 3.3 kHz in human and 4.0 kHz in monkey (macaque).Frequency selectivity and neural phase-locking can be characterized from mass potentials with a minimally invasive technique from awake, normal hearing subjects. We found evidence that a) humans have sharper frequency tuning than commonly studied animals, consistent with previous reports (Shera et al., 2002; Oxenham and Shera, 2003) and; b) humans have an upper frequency limit of phase-locking that is not higher than in cat (5 kHz), and likely is somewhat lower.
Choose an application
Choose an application
Choose an application
The largest nucleus of the human superior oliva ry complex is the medial superior olive (MSO), whi ch generates sensitivity to interaural time differ ences (ITD), the main cue for horizontal sound loc alization, by comparing the inputs it receives fro m both ears. The MSO is dysmorphic in autistic spe ctrum disorders, corresponding to the frequent occ urrence of auditory dysfunction in these patients. The computation resulting in sensitivity to ITD is thought to be coincidence detection, a n operation that allows neurons to decode time cod es, but it has not been studied directly in vivo d ue to the absence of intracellular recor dings. MSO neurons are tuned to differen t best ITDs, which means that there should be an internal delay in the brainstem compensat ing for the external ITD. The source of internal delay has been a controversial issue . MSO responses to stimuli beyond tones are a lmost completely absent from the literature. Besid es excitatory inputs, MSO neurons receive inhibito ry inputs from each ear, from the latera l nucleus of the trapezoid body (LNTB; ipsilateral ly) and the medial nucleus of the trapezoid b ody (MNTB; contralaterally). Especially the proper ties of the LNTB input are unclear, due to th e lack of in vivo recordings from retrie ved cells.We performed in vivo whole ce ll recordings from MSO and LNTB neurons in th e Mongolian gerbil, and labeled these neurons ¨with biocytin. Intracellular recordings during mo naural and binaural sounds (tones and noise) allow ed us to separate subthreshold (input) from s uprathreshold (output) activity and study the ¨neural computation performed. Pharmacological man ipulation were used to isolate excitatory and ¨inhibitory inputs. Finally, we performed a c oincidence analysis of broadband responses of ¨cat trapezoid body fibers to study the effect of¨ number of inputs, binaural and monaural¨ coincidence threshold and coincidence window on ps eudobinaural functions.Intracellular recordi ngs from MSO neurons show that these neurons mostly receive excitatory postsynaptic potentials (EPSP s) and only few and small inhibitory pos tsynaptic potentials (IPSPs). Pharmacological mani pulation revealed that ITD tuning of the neurons is not systematically affected by inhibit ion, in contrast with an earlier hy pothesis. Instead, we find a shift between the summation of subthreshold inputs and suprat hreshold ITD tuning, due to subtle asymmetrie s in the temporal pattern of EPSPs that¨ differentially activate voltage-gated potassi um channels, providing an unexpected source of internal delay. Responses from tones at dif ferent frequencies and noise bandwidths show¨ an impressive boost of MSO output to low ¨frequency tones and narrowband noise, due to decr easing coherence of input spikes for higher f requencies and broader bandwidths. Coincidence cou nts of responses of trapezoid body fibers result i n pseudobinaural noise delay functions and rate-co rrelation functions. We show that multiple in puts are needed per side to get ade quate output spike rates; that monaural coinc idences have to be suppressed over binau ral ones to maintain binaural sensitivity and that the neuron needs to be sensitive to sin gle coinciding spikes. We find anatomical evidence ¨for at least two subdivisions in the LN TB, termed pv (posteroventral) LNTB and m (main) L NTB. These divisions also have different phys iological properties. We find an unexpected projec tion from the lateral superior olive (LSO) to the¨ LNTB.We conclude that the MSO performs adapt ive coincidence detection rather than instant aneous coincidence detection and is sensitive to a ¨low number of input spikes. In addition to i ts binaural properties, MSO neurons are well place d to detect low frequency sounds. The LNTB is ¨composed of several subdivisions that can be ¨distinguished anatomically and physiologically, a nd receives an unexpected input from the ¨LSO.Publication status: acceptedKU Leuven publication type: THNumber of times cited:0Appears in Collections:Laboratory for Auditive NeurophysiologyResearch Group NeurophysiologyFiles in This Item:There are no files associated with this item.
Choose an application
Luisteren met twee oren heeft enorme voordelen t.o.v. luisteren met één oor. Eén van de voordelen is het vermogen om laagfrequente geluiden ruimtelijk te lokaliseren. Dit kunnen we door onze gevoeligheid voor kl eine interaurale tijdsverschillen (interaural time differences, ITDs) in de geluiden die de twee oren bereiken. Het doel van deze thesis is om m eer inzicht te krijgen in de neurale mechanismen die deze ITD gevoelighe id onderbouwen. Geluiden in beide oren worden temporeel gecodeerd via he t proces van fasekoppeling of fasebinding. Deze temporele code wordt via de detectie van coïncidenties in de mediale bovenste olijfkern (medial superior olive, MSO) omgezet tot een gevoeligheid voor ITD. In deze thes is bestuderen we opeenvolgende stappen in de neurale banen voor lokalisa tie van laagfrequente geluiden. We registreerden de neurale activiteit v an individuele neuronen op monauraal niveau: gehoorzenuw (auditory nerve , AN), corpus trapezoideum (trapezoid body, TB), en de mediale nucleus h iervan (medial nucleus of the trapezoid body, MNTB), alsook op binauraal niveau: van individuele neuronen in de colliculus inferior (IC). We mat en ook lokale veldpotentialen in de MSO. Onze specifieke vragen groepere n we in drie onderwerpen: 1) Vergelijking van monaurale en binaurale stadia in de hersenstam Voor monaurale taken zoals frequentiediscriminatie zijn smalle auditieve filters voordelig. Voor sommige binaurale taken zouden brede filters ec hter nuttiger zijn. We vergeleken de bandbreedte (bandwidth, BW) van mon aurale vezels in de AN en TB met binaurale neuronen in de IC en vonden d at ze vergelijkbaar waren, wat er op duidt dat de bandbreedte van de coc hleaire filters bewaard blijft op hogere niveaus. De gevoeligheid voor c orrelatie verschilt echter: AN vezels zijn zeer gevoelig voor veranderin g in correlatie en deze gevoeligheid neemt toe in de TB. Deze laatste ve zels lijken gespecialiseerd te zijn om binaurale temporele vergelijkinge n te verbeteren. Tegen de verwachtingen in blijkt dat de gevoeligheid vo or correlatie gedegradeerd is op niveau van de IC. Dit betekent dat de t emporele informatie die beschikbaar is op het monaurale niveau niet ten volle benut wordt door het binaurale systeem, en kan een indicatie zijn dat vroegere veronderstellingen over de precisie van het coïncidentie-de tectie process in de MSO incorrect zijn. 2) De getrouwheid van synaptische transmissie in de MNTB TB vezels geven input aan de MNTB via de grootste synaps in de hersenen: de calyx van Held. De MNTB is inhibitorisch op de MSO, en er zijn bewer ingen dat temporeel precieze inhibitorische input noodzakelijk is voor I TD gevoeligheid. Anderzijds is er recente evidentie die suggereert dat s ynaptische transmissie t.h.v. de calyx van Held kan falen. We tonen dat in de kat de synaps toch betrouwbaar is in termen van het aantal actiepo tentialen dat doorgegeven wordt, maar minder in termen van het tijdstip van afvuren van die actiepotentialen.De grootte en specialisatie van de calyx van Held in acht nemende, geven deze resultaten een indicatie dat het zinvol is om de wijze waarop andere, kleinere synapsen hun temporele informatie doorgeven in vraag in te stellen. Onze resultaten toonden aa n dat een verhoging in het aantal actiepotentialen (i.e. in stimulus int ensiteit) gedragsmatig belangrijke vertragingen kunnen veroorzaken. Het aantal afgevuurde actiepotentialen en de latentie van neurale antwoorden vergroten met stimulus intensiteit in het perifere auditieve systeem, n et zoals in andere sensoriële systemen. Beide factoren, synaptische vert ragingen en neurale latentie, werken elkaar tegen. De interactie tussen deze beide delays bij het veranderen van stimulus intensiteit vereist ve rder onderzoek. 3) De bron van de neurophonic in de MSO Het Jeffress model van binauraal horen postuleert dat er een ruimtelijke kaart van ITDs is in de MSO. Het registreren van individuele MSO neuron en is moeilijk vanwege een sterke extracellulaire veldpotentiaal gekend als de neurophonic. Deze veldpotentiaal werd bestudeerd in vogels: ver tragingen in de potentiaal werden waargenomen en werden geïnterpreteerd als evidentie voor de ruimtelijke kaart gepostuleerd door Jeffress. We r egistreerden de neurophonic in de kat en vonden vertragingen die erg lij ken op die in vogels. We tonen echter dat de vertragingen in de kat het resultaat zijn van dipolen in de MSO, en dus niet noodzakelijk op een ru imtelijke kaart wijzen georienteerd volgens de korte as van de MSO. Het begrijpen van de oorsprong van deze grote vertragingen in de dipolen ste lt ons in staat om op zoek te gaan naar een ruimtelijke kaart van ITDs g eorienteerd volgens de lange as van ...
Choose an application
Choose an application
Forensic science involves the application of a broad spectrum of scientific principles within a legal framework, and is typically used to support criminal investigations. Drawing from various scientific fields such as physics, chemistry, and biology, forensic science is a vital tool in unearthing the truth in any legal proceeding. Over the course of its existence, forensic science evolved towards an evidence-based framework, relying on rational procedures. Consequently, today's focus of forensic investigators is on the recognition, identification, and analysis of evidence, all made possible through the introduction of countless techniques and technological advances. However, in spite of all technological advances already made, forensic investigations are not solved by the push of a button. Crime laboratories employ large teams of investigators that often have to perform tedious and time-consuming tasks. Such manual work is detrimental to the overall investigation, as it introduces subjectivity into the analysis, which in turn may lead to miscarriages of justice. Moreover, having forensic investigators perform lengthy analyses on a crime scene increases the risk of introducing, removing, or altering crucial evidence. Finally, considering the fact that forensic investigators are highly skilled experts, manual work increases the cost of the overall investigation.In this thesis, our main aim is therefore on replacing time-consuming work by automated procedures in an attempt to tackle the aforementioned problems. More specifically, our focus is on the application of data-driven image computing methods to further advance the automation in the fields of Bloodstain Pattern Analysis and Virtual Autopsy.In the first part of this thesis, we focused on the field of Bloodstain Pattern Analysis. Concretely, we developed a software package called HemoVision, which enables fully automated analysis of impact patterns. In doing so, three major contributions were made. First, we developed a statistical shape model that enables fully automated analysis of individual bloodstains. Compared to existing methods, the proposed approach is faster, more accurate, and more objective. Second, we proposed a vision-based approach using fiducial markers to automatically reconstruct impact patterns from various images. This step was crucial to the development of HemoVision, as it removes most of the manual work that is required with existing methods. Third, we integrated the proposed methods in an intuitive interface, allowing forensic investigators to use our software. Moreover, we performed a validation study to asses HemoVision's overall accuracy. Compared to the state-of-the-art, HemoVision obtains competitive results.In the second part of this thesis, we turned to the field of Virtual Autopsy. More specifically, our focus was on the development of a general framework that enables the automatic detection of abnormalities in medical images. Again, three major contributions were made. First, a data-driven preprocessing method for medical images was developed, allowing to correct imbalances in tissue intensity distribution widths. This is an important step, as standard image processing methods are negatively influenced by this so-called heteroscedasticity. Second, we developed a semi-supervised exemplar-based method to detect abnormalities in medical images. In contrast to existing approaches, the proposed method does not require annotated training data, does not rely on ill-posed non-rigid deformations, and is independent of the imaging modality. Third, we applied the aforementioned methods to the problem of automatic gunshot trajectory reconstruction. Using a robust linear regression model, we were able to correctly estimate trajectories in the majority of the cases. To the best of our knowledge, this is the first method that allows fully automatic reconstruction of gunshot trajectories.Overall, we believe we have contributed significantly to the fields of Bloodstain Pattern Analysis and Virtual Autopsy. By introducing quantitative, automated methods, forensic investigations become less costly, more objective, and faster, which in the end, is beneficial to all people involved in criminal investigations.
Choose an application
Choose an application
De mediale olivocochleaire baan is een efferente zenuwbaan die cellen gelegen rond de mediale oliva superior in de hersenstam met de buitenste haarcellen in het slakkenhuis verbindt. Door deze baan elektrisch of akoestisch te activeren, vindt de mediaal olivocochleaire reflex plaats. In de afwezigheid van de mediaal olivocochleaire reflex staan de buitenste haarcellen in het slakkenhuis in voor cochleaire versterking. Dit actief mechanisme van de buitenste haarcellen zorgt voor een lagere gehoordrempel, vergroot dynamisch bereik en verbeterde frequentieselectiviteit. In de aanwezigheid van de mediaal olivocochleaire reflex wordt de activiteit van de buitenste haarcellen in het slakkenhuis onderdrukt. Hierdoor zal de gehoordrempel toenemen, het dynamische bereik verkleinen en de frequentieselectiviteit verminderen. Bij dieren vonden er reeds uitgebreide onderzoeken plaats en de effecten van de mediale olivocochleaire reflex redelijk goed begrepen. Bij mensen is dit onderzoek echter minder evident, aangezien het uitvoeren van directe experimenten omwille van ethische redenen niet mogelijk is. Dit heeft tot gevolg dat er anno 2017 nog steeds meerdere onbeantwoorde vragen zijn, met betrekking tot de activatie en de functie van de mediaal olivocochleaire reflex bij mensen. Huidig onderzoek bestond uit een reeks van psychofysische metingen bij normaalhorende volwassenen. Op basis van deze testen is getracht de mediaal olivocochleaire reflex op een indirecte manier te activeren door het aanbieden van contralaterale geluidsstimuli. De resultaten gaven aan dat door een contralateraal breedbandig ruissignaal aan te bieden, voornamelijk de lagere frequenties aan het ipsilaterale oor moeilijker hoorbaar worden. Dit in tegenstelling met de hogere frequenties, waarbij de drempelniveaus relatief ongewijzigd blijven. Daarnaast gaven de resultaten aan dat de bandbreedte van de contralaterale geluidsstimuli een rol speelt in de hoorbaarheid van een toon. Zo zorgt een kleinere bandbreedte voor een minder groot effect van de mediaal olivocochleaire reflex. Tot slot lijkt de centerfrequentie van de contralaterale geluidstimuli geen noemenswaardige rol te spelen. Al is het effect van een contralateraal ruissignaal bestaande uit lagere frequenties groter dan wanneer hetzelfde ruissignaal voornamelijk hoge frequenties bevat. Omwille van de beperkte groepsgrootte is verder onderzoek naar het mediaal olivocochleaire systeem bij mensen aangewezen.
Listing 1 - 10 of 16 | << page >> |
Sort by
|