SEARCH · PubMed Health

Results for “threshold model”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,729 records · Page 96Linked to original sources

PubMed2025

Donor-Derived Cell-Free DNA Stratifies Risk of Mortality and Graft Dysfunction in Severe Acute Cardiac Allograft Rejection.

BACKGROUND: Cardiac acute rejection (AR) is a risk factor for poor outcomes, however there are limited risk prediction models to stratify patients for death or sustained LV dysfunction. This study assesses the prognostic utility of percentage donor-derived cell-free DNA (%dd-cfDNA) at the diagnosis of AR for poor outcomes. METHODS: The prospective multicenter GRAfT study enrolled heart transplant recipients and collected serial plasma samples to quantitate %dd-cfDNA. AR was defined as acute cellular rejection (ACR), antibody-mediated rejection (AMR), as well as biopsy-negative AMR (donor-specific antibody positivity with LV dysfunction). AR was classified as mild-to-moderate (ACR grade 2 or AMR grade 1) or severe (ACR grade ≥3, AMR grade ≥2, or DSA+/LV dysfunction) and further stratified by a %dd-cfDNA threshold of 0.25%. Regression models assessed the association between AR and %dd-cfDNA levels at the AR diagnosis with the primary composite outcome of sustained LVEF decline <50% and/or death. RESULTS: The study included 275 patients and 3,190 %dd-cfDNA assessments. Over the median of 4.6 (IQR 1.8 - 5.0) years follow-up, 51 patients experienced the composite outcome of death or prolonged EF reduction, and 75 patients developed AR, including 16.2% patients with ACR, 9.4% with pathologic AMR, and 6.6% with DSA+/LV dysfunction. Thirty-two (42.7%) patients had severe AR and 43 (57.3%) had mild-to-moderate AR. Severe-but not mild-to-moderate-AR was associated with an increased risk of the primary composite endpoint (HR = 5.17, 95% CI 2.38 - 11.3, p < 0.001). Among those with severe AR, a %dd-cfDNA level greater than 0.25% at diagnosis was associated with a higher risk of the primary outcome (HR, 6.06, 95% CI, 1.78-20.6; p = 0.004). Percent dd-cfDNA remained elevated in severe AR patients with adverse outcomes. CONCLUSION: Severe AR with high %dd-cfDNA levels is associated with an increased risk of poor outcomes, offering novel prognostic utility.

Acute Rejection↗

PubMed1991

A statistical method for assessing a threshold in epidemiological studies.

I describe a method for estimating and testing a threshold value in epidemiological studies. A threshold effect indicates an association between a risk factor and a defined outcome above the threshold value but none below it. An important field of application is occupational medicine where, for a lot of chemical compounds and other agents which are non-carcinogenic health hazards, so-called threshold limit values or TLVs are specified. The method is presented within the framework of the logistic regression model, which is widely used in the analysis of the relationship between some explanatory variables and a dependent dichotomous outcome. In most available programs for this and also for other models the concept of a threshold is disregarded. The method for assessing a threshold consists of an estimation procedure using the maximum-likelihood technique and a test procedure based on the likelihood-ratio statistic R, following under the null hypothesis (no threshold) a quasi one-sided chi 2 distribution with one degree of freedom. This use of this distribution is supported by a simulation study. The method is applied to data from an epidemiological study of the relationship between occupational dust exposure and chronic bronchitic reactions. The results are confirmed by bootstrap resampling.

Bronchitis↗

PubMed1997

A model of Weber and noise gain control in the retina of the toad Bufo marinus.

We present several variations of a model of gain control in the retina of the toad Bufo marinus, and use them to fit the threshold-vs-intensity data of an actual toad ganglion cell [Donner et al. (1990). Journal of General Physiology, 95, 733-753]. Our models are based on a proposal by Donner et al. that the gain (neural spike per photon ratio) of toad ganglion cells is set by a sequence of two retinal gain control stages. The first stage consists of a Weber gain control mechanism at the level of the red rods. The second is a more proximal "noise gain" stage, which multiplies the (incremental) input signal by a factor that is inversely proportional to the standard deviation of the random ganglion cell input and, under conditions that produce the de Vries-Rose threshold law, is also proportional to the standard deviation of the photon fluctuations within the ganglion cell receptive field. We demonstrate that noise gain control arises naturally from modeling ganglion cell spike generation with either of two common types of spike generation models: integrate-and-fire models or threshold accommodation models. We simulate the process of spike generation in both types of models and show that either model can account for the basic overall shape of the toad t.v.i. curve. However, although integrate-and-fire models appropriately generate noise gain control, they cannot quantitatively fit the threshold data with realistic retinal parameters. Integrate-and-fire models also fail to account for the observed relationship between the generator potential of the ganglion cell and its spiking probability. A threshold accommodation model with realistic retinal parameters, on the other hand, can account for both the threshold data and the generator potential-spike probability relationship. When a Weber gain stage is added to the model at the photoreceptor level, the resulting two-stage gain control model is shown to account quantitatively for the ganglion cell t.v.i. curve of Bufo marinus over the full range of background levels studied by Donner et al.

Adaptation, Ocular↗

PubMed1988

Resonance-frequency discrimination.

Measurements of the just-noticeable change in resonance frequency delta Fr of a second-order filter are reported. The source signal was either periodic, with or without a smooth change in fundamental frequency, or it was random white noise. These differences in the nature of the source had little effect on delta Fr. Over the investigated ranges of reference resonance frequency (Fr = 300 to 2000 Hz) and filter selectivity (Q = 1 to 36), the results are well summarized by delta Fr = 0.079 Fr/square root Q. The data were used to evaluate filter-bank models employing different filter shapes and performance-prediction schemes. A good fit to the resonance-discrimination data was obtained with filters derived in a masking study by Patterson [J. Acoust. Soc. Am. 55, 802-809 (1974)] and with performance based on use of the maximum level difference over all bands. The latter finding indicates that listeners may not make optimal use of small level differences distributed over multiple bands.

Acoustic Stimulation↗

PubMed2001

Sample size requirement for detecting interference under the chi-square model.

The chi-square model (CHS) of recombination has been studied extensively, in recent years, for its ability of capturing the process and estimating the level of crossover interference. Exploration thus far shows that this model yields much better fits to human genetic data than Haldane's no-interference model, and the explicit level of interference can be easily estimated as well. This paper provides calculations of sample sizes required to detect interference under CHS for a variety of settings. Two data types, fully informative meioses and phase-unknown backcross families, are studied. Under each setting, we calculate the number of meioses/families needed to ensure that the expected log-likelihood difference between the chi(2) interference model and Haldane's no-interference model exceeds a prespecified threshold. It is found that joint consideration of multiple (more than three) markers dramatically reduces the number of meioses/families needed when compared to an analysis based on three-point data, a traditional setting for detecting interference using three-point tests. The results indicate that the numbers of meioses needed to detect interference under CHS are well within the reach of most genetic mapping studies.

Chi-Square Distribution↗

PubMed2001

Bcl-2 expression as a novel immunohistochemical marker for ruptured tubal ectopic pregnancy.

Programmed cell death by apoptosis occurs in fetal and maternal tissues during early pregnancy and plays an important role during implantation, decidualization, and in fetal development. In the regulation of apoptosis, bcl-2 is one of the central controlling genes, and acts by protecting the cell against apoptosis. It is postulated that invasiveness of ectopic trophoblast towards and through the muscularis zone of the tubal wall consequently leading to tubal rupture might be due to disturbed regulation of apoptosis. By means of immunohistochemistry and a computerized image analysis, bcl-2 immunostaining was localized and quantified in 36 randomly selected paraffin-embedded ectopic trophoblast tissue specimens collected from women undergoing surgery for ruptured (n = 18) and non-ruptured (n = 18) tubal ectopic pregnancies. Immunostaining was found in the villi syncytiotrophoblast in all patients, while the percentage of positive bcl-2 immunostained area (%PA) (P = 0.0009) and staining intensity (P = 0.0042) were consistently greater in the group of ruptured ectopic pregnancies. Including the variables %PA and saturation into a logistic regression model for a probability threshold of 0.5 (<0.5 = non-ruptured ectopic pregnancy, >0.5 = ruptured ectopic pregnancy) to identify tubal rupture, a sensitivity and specificity of 94.4% were found. It is suggested that elevated bcl-2 immunostaining in the syncytiotrophoblast layer reflects unlimited cell survival of ectopic trophoblast and could lead to the establishment of a circulating marker for tubal rupture.

Adult↗

PubMed1994

Masked detection thresholds and temporal integration for noise band signals.

Masked detection thresholds were measured at a center frequency of 2500 Hz for a range of noise signal bandwidths (W = 62 to 6000 Hz) and durations (T = 10 to 480 ms) approximating that found acoustically in speech. The signals were presented in an uncorrelated 480-ms, 6000-Hz-wide masker. The masker was presented: (1) at a constant spectrum level (53 dB SPL/Hz) or (2) with the overall level varied randomly over a 50-dB range from interval to interval of a trial. Performance was disrupted in the random-level masker only for the condition where the signal and uncorrelated masker were gated on and off simultaneously and were matched spectrally. Time constants (tau) estimated from temporal integration functions fit to the masked detection threshold data were related inversely to W for W broader than the critical bandwidth. Sensitivity to the noise signals was evaluated in the context of an optimum-detection model (Green, 1960). The results did not follow the prediction of a constant bandwidth-duration (WT) product, but may be understood in terms of cues available to the listener from the relative combination of signal and masker parameters. At least three cues for detection were identified in these experiments: (1) a relative timing cue, (2) a spectral shape cue, and (3) a traditional energy cue compared across observation intervals. The relative timing cue and spectral shape cue together contributed as much as a 10- to 12-dB advantage relative to detection based on the traditional energy cue alone. A new multi-cue detection model for predicting the masked detection thresholds was derived. Predictions from the new model were highly correlated (r = 0.95) with the empirically measured masked detection thresholds.

Adult↗

PubMed1981

Acoustic pressure amplitude thresholds for rectified diffusion in gaseous microbubbles in biological tissue.

One of the mechanisms often suggested for the biological action of ultrasonic beams irradiating human tissues is concerned with the presence in the tissues of minute gaseous bubbles which may, under the influence of the ultrasonic field be stimulated to grow to a size a which resonance or collapse occurs with severe associated shear stresses. The evidence for the existence of microbubbles in tissues is reviewed. The results of calculations, using two existing theoretical models, of the peak pressure threshold as a function of frequency are presented. The frequency is normalized with the resonant frequency of the bubble, and results are presented for three bubble radii (1, 2, and 3.5 micrometer) and for different values of the gas concentration in the tissue between 0.1 and 1. The results from two models differ suggesting that an improved model and better experimental data for the threshold calculations would be appropriate for further calculations. The thresholds calculated range below the peak pressure amplitudes used in continuous wave diagnostic instruments, indicating the need for a more careful investigation both of this damage mechanism and of the exposures used in routine diagnosis. The results of calculations for typical (transient) exposure conditions from pulse-echo equipment are presented, indicating that rectified diffusion and stable cavitation are improbable phenomena in these circumstances.

Animals↗

PubMed2000

Modeling place fields in terms of the cortical inputs to the hippocampus.

A model of place-cell firing is presented that makes quantitative predictions about specific place cells' spatial receptive fields following changes to the rat's environment. A place cell's firing rate is modeled as a function of the rat's location by the thresholded sum of the firing rates of a number of putative cortical inputs. These inputs are tuned to respond whenever an environmental boundary is at a particular distance and allocentric direction from the rat. The initial behavior of a place cell in any environment is simply determined by its set of inputs and its threshold; learning is not necessary. The model is shown to produce a good fit to the firing of individual place cells, and populations of place cells across environments of differing shape. The cells' behavior can be predicted for novel environments of arbitrary size and shape, or for manipulations such as introducing a barrier. The model can be extended to make behavioral predictions regarding spatial memory.

Afferent Pathways↗

PubMed1995

Evaluation of prostate-specific antigen as a surrogate marker for response of hormone-refractory prostate cancer to suramin therapy.

PURPOSE: We evaluated the surrogate role of serum prostate-specific antigen (PSA) using prospectively collected information from patients with hormone-refractory prostate cancer (HRPC) treated with suramin. MATERIALS AND METHODS: Data from 103 patients were analyzed using survival analysis, exploratory analysis, and regression analysis. RESULTS: There was a significant survival difference between groups of patients with a PSA decrease of < or = 0% or greater than 0% (P = .018). There were no significant overall survival differences between groups of patients with PSA decreases less than 50% or > or = 50% and less than 75% or > or = 75%. Tree-based modeling did not define a specific threshold percentage PSA change as a response criterion. For a response of 1-year survival, sensitivity increased (0.91 v 0.69), but specificity decreased (0.37 v 0.62), with a 75% versus 50% PSA decrease used as classification criterion. Differences between the area under the receiver-operating curves (ROCs) with 50% and 75% PSA decreases as threshold values were small. For a response of 1-year survival, attributable proportions were 0.38 and 0.68, respectively, with 50% and 75% PSA decreases as threshold values. When pretreatment variables were assessed by Cox proportional hazards model, hemoglobin level was the most significant predictor of survival. When percentage PSA change was included in the model, hemoglobin level remained the most significant factor, but percentage PSA change was also a weak, but statistically significant, factor. PSA was a weak, but statistically significant, predictor of survival in Cox proportional hazards model with PSA as a time-variant covariate. CONCLUSION: Reduction in PSA level has weak prognostic significance with respect to survival in HRPC patients, but, currently, PSA reduction cannot be used as a reliable response criterion to evaluate treatment efficacy in individual patients. Prospective, randomized studies, including prospective measurement of other indices related to symptomatic clinical benefits, are required.

Adult↗

PubMed2000

Visual-field defects in well-defined retinal lesions using Humphrey and Dicon perimeters.

BACKGROUND: This study was undertaken to correlate field defects in patients having well-defined retinal lesions within the central 30 degrees using Humphrey and Dicon perimetry standard threshold programs and to compare these results against the standard of fundus photography. METHODS: Eleven eyes of 11 patients had various well-defined retinal lesions. Ten of those subjects had best-corrected visual acuity of 20/25 or better. Subjects had either one previous field or no previous experience with any automated perimeter and were tested using either the 30-2 Full Threshold or SITA Standard program of the Humphrey Field Analyzer (HFA) Model 750 and the Central Grid Threshold program (#9) (a full-threshold program) of the Dicon LD-400 perimeter. Defects were scored and compared with fundus photographs scored by an independent observer. RESULTS: Significant high positive correlation coefficients were found between Humphrey, Dicon, and fundus photography, for mean number of significant field defects and for mean deviations. There were no significant differences between the mean scores, mean deviations, or fixation loss percentages. The time the patient occupied a perimeter was 56% less with the Dicon perimeter than with the Humphrey perimeter when the Full Threshold test was used and 8% less when the SITA Standard was used. CONCLUSION: In this study of patients with well-defined retinal lesions, both Humphrey and Dicon Central Threshold field programs exhibited an equally high positive correlation in mapping scotomas of expected size and depth when compared with fundus photography, with no significant difference in fixation loss frequency. However, the total testing time was less with Dicon perimetry than with Humphrey perimetry.

Adolescent↗

PubMed2000

Detection of quasitrapezoidal frequency and amplitude modulation.

It has been proposed that the detection of frequency modulation (FM) of sinusoidal carriers can be mediated by two mechanisms; a place mechanism based on FM-induced amplitude modulation (AM) in the excitation pattern, and a temporal mechanism based on phase locking in the auditory nerve. The temporal mechanism appears to be "sluggish" and does not play a role for FM rates above about 10 Hz. It also does not play a role for high carrier frequencies (above about 5 kHz). This experiment provided a further test of the hypothesis that the effectiveness of the temporal mechanism depends upon the time spent close to frequency extremes during the modulation cycle. Psychometric functions for the detection of AM and FM were measured for two carrier frequencies, 1 and 6 kHz. The modulation waveform was quasitrapezoidal. Within each modulation period, P, a time Tss was spent at each extreme of frequency or amplitude. The transitions between the extremes, with duration Ttrans had the form of a half-cycle of a cosine function. The modulation rate was 2, 5, 10, or 20 Hz, giving values of P of 500, 200, 100, and 50 ms. TSS varied from 0 ms (sinusoidal modulation) up to 160, 80, 40, or 20 ms, for rates of 2, 5, 10, and 20 Hz, respectively. The detectability of AM was not greatly affected by modulation rate or by the value of TSS, except for a slight improvement with increasing TSS for the lowest modulation rates; this was true for both carrier frequencies. For FM of the 6-kHz carrier, the pattern of results was similar to that found for AM, which is consistent with an excitation-pattern model of FM detection. For FM of the 1-kHz carrier, performance improved markedly with increasing TSS, especially for the lower FM rates; there was no change in performance with TSS for the 20-Hz modulation rate. The results are consistent with the idea that detection of FM of a 1-kHz carrier is partly mediated by a sluggish temporal mechanism. That mechanism benefits from greater time spent at frequency extremes of the modulation cycle for rates up to 10 Hz.

Auditory Perception↗

PubMedDate not supplied

Insertion torque and resonance frequency analysis of dental implant systems in an animal model with loaded implants.

PURPOSE: The aim of this study was to compare insertion torque and resonance frequency analysis of different implant systems in an animal model with loaded implants. MATERIALS AND METHODS: Three types of Brånemark implants (machined MkIII, TiUnite MkIII, and MkIV) and 2 types of Straumann implants (sandblasted, large-grit, acid-etched [SLA] and titanium plasma-sprayed [TPS]) were studied. Thirty-two implants of each type (n = 160) were placed in 16 beagle dogs. Maximum insertion torque values were recorded. After a healing period of 8 weeks, the implants were loaded for 3 months; the animals were then sacrificed. At placement, after healing, and at the end of the loading phase, resonance frequency analysis was performed and implant stability quotients (ISQs) were recorded. RESULTS: Higher insertion torque values were seen for the conical MkIV than for the MkIII. No difference was seen between the Brånemark and Straumann implants on the basis of ISQ value at placement. ISQ and insertion torque values were lower for the cylindric Straumann implants than for the self-tapping implants. For all implant systems a significant decrease in median ISQ was observed, with a median decrease ranging from 3 to 6. ISQ values for self-tapping implants remained stable after loading, whereas the ISQ values for non-self-tapping cylinders decreased. The maximum insertion torque values for failed and successful implants were not significantly different. Significantly higher ISQ values at placement were seen for successful implants (P =.003). Based on this model for ISQ, a threshold of 65.5 was identified, with a sensitivity of 83% and specificity of 61% for prediction of implant loss. ISQ values at the start of loading were not predictive of implant loss in the loading period. CONCLUSION: Caution should be used when judging implant systems on the basis of resonance frequency analysis and torque measurement.

Animals↗

PubMed2004

A role for contrast-normalisation in second-order motion perception.

Two low-level motion models are applied to a second-order stimulus, a translating contrast modulation of static binary noise. Both models have been used to demonstrate equivalence between energy and gradient algorithms and can be split into a motion-opponent stage followed by a contrast-normalised stage. Analysis of results shows no directional bias at the motion-opponent stage but a strong bias, indicating the correct direction of second-order motion, at the contrast-normalised stage. This demonstrates that the intrinsically non-linear process of contrast-normalisation may play a part in the detection of second-order motion.

Algorithms↗

PubMed2001

Susceptibility differences in chemical carcinogenesis linearize the dose-response relationship: threshold doses can be defined only for individuals.

The debate on thresholds in dose-response relationships for chemical carcinogenesis concentrates on the question of mechanisms of action that come into play only at dose levels that overwhelm compensatory control mechanism, such as DNA repair or regulation of cell proliferation and death. In this article, individual susceptibility differences are introduced. It is postulated that one single threshold dose cannot be defined for a heterogeneous population because both the background rate of carcinogenesis and specific exposure-related effects differ between individuals. A threshold dose can therefore be defined only on an individual basis and for a given organ. Expressed as a time-to-tumor, the threshold dose results in tumor manifestation at exactly the end of the specified observation period. For those individuals who do not have cancer by the end of this period, the dose was below their individual threshold dose, for those who do have cancer, the dose was above their threshold dose. Based on this concept, a distinction between genotoxic and non-genotoxic carcinogens is no longer required; both types modulate time-to-tumor. Although the present analysis does not allow to define a threshold dose for a population, the setting of a "limit value" for regulatory purposes can be considered if regulators are aware of the fact that this splits a population at some percentile into a group for which the chosen standard is protective and a group for which it might not be. Investigation of factors that confer particular susceptibility to individuals is the key to an understanding of the dose-response relationship at low dose.

Animals↗

PubMed2000

Visual contributions to human self-motion perception during horizontal body rotation.

It is still an enigma how human subjects combine visual and vestibular inputs for their self-motion perception. Visual cues have the benefit of high spatial resolution but entail the danger of self motion illusions. We performed psychophysical experiments (verbal estimates as well as pointer indications of perceived self-motion in space) in normal subjects (Ns) and patients with loss of vestibular function (Ps). Subjects were presented with horizontal sinusoidal rotations of an optokinetic pattern (OKP) alone (visual stimulus; 0.025-3.2 Hz; displacement amplitude, 8 degrees) or in combinations with rotations of a Bárány chair (vestibular stimulus; 0.025-0.4 Hz; +/- 8 degrees). We found that specific instructions to the subjects created different perceptual states in which their self-motion perception essentially reflected three processing steps during pure visual stimulation: i) When Ns were primed by a procedure based on induced motion and then they estimated perceived self-rotation upon pure optokinetic stimulation (circular vection, CV), the CV has a gain close to unity up to frequencies of almost 0.8 Hz, followed by a sharp decrease at higher frequencies (i.e., characteristics resembling those of the optokinetic reflex, OKR, and of smooth pursuit, SP). ii) When Ns were instructed to "stare through" the optokinetic pattern, CV was absent at high frequency, but increasingly developed as frequency was decreased below 0.1 Hz. iii) When Ns "looked at" the optokinetic pattern (accurately tracked it with their eyes) CV was usually absent, even at low frequency. CV in Ps showed similar dynamics as in Ns in condition i), independently of the instruction. During vestibular stimulation, self-motion perception in Ns fell from a maximum at 0.4 Hz to zero at 0.025 Hz. When vestibular stimulation was combined with visual stimulation while Ns "stared through" OKP, perception at low frequencies became modulated in magnitude. When Ns "looked" at OKP, this modulation was reduced, apart from the synergistic stimulus combination (OKP stationary) where magnitude was similar as during "staring". The obtained gain and phase curves of the perception were incompatible with linear systems prediction. We therefore describe the present findings by a non-linear dynamic model in which the visual input is processed in three steps: i) It shows dynamics similar to those of OKR and SP; ii) it is shaped to complement the vestibular dynamics and is fused with a vestibular signal by linear summation; and iii) it can be suppressed by a visual-vestibular conflict mechanism when the visual scene is moving in space. Finally, an important element of the model is a velocity threshold of about 1.2 degrees/s which is instrumental in maintaining perceptual stability and in explaining the observed dynamics of perception. We conclude from the experimental and theoretical evidence that self-motion perception normally is related to the visual scene as a reference, while the vestibular input is used to check the kinematic state of the scene; if the scene appears to move, the visual signal becomes suppressed and perception is based on the vestibular cue.

Adult↗

PubMed1999

Immigration and the Maintenance of Local Species Diversity.

Explaining the maintenance of high local species diversity in communities governed by competition for space has been a long-standing problem in ecology. We present a simple theoretical model to explore the influence of immigration from an external source on local coexistence, species abundance patterns, and ecosystem processes in plant communities. The model is built after classical metapopulation models but is applied to competition for space between individuals and includes immigration by a propagule rain and an extinction threshold for rare species. Our model shows that immigration can have a huge effect on local species diversity in competitive communities where competition for space would lead to the exclusion of all but one species if the community were closed. Local species richness is expected to increase strongly when immigration intensity increases beyond the threshold required for the successful establishment of one or a few individuals. Community structure and species relative abundances are also expected to change markedly with immigration intensity. Increasing immigration causes total space occupation by the community to increase but primary productivity on average to either decrease or stay constant with increasing diversity, depending on the relation between immigration and local reproduction rates. These results stress the need for a regional perspective to understand the processes that determine species diversity, species abundance patterns, and ecosystem functioning in local communities.

competition for space↗

PubMed1999

Computer models for masked hearing experiments with beluga whales (Delphinapterus leucas).

Environmental assessments of manmade noise and its effects on marine mammals need to address the question of how noise interferes with animal vocalizations. Seeking the answer with animal experiments is very time consuming, costly, and often infeasible. This article examines the possibility of estimating results with software models. A matched filter, spectrogram cross-correlation, critical band cross-correlation, and a back-propagation neural network detected a beluga vocalization in three types of ocean noise. Performance was compared to masked hearing experiments with a beluga whale [C. Erbe and D. M. Farmer, Deep-Sea Res. II 45, 1373-1388 (1998)]. The artificial neural network simulated the animal data most closely and raised confidence in its ability to predict the interference of a variety of noise source with a variety of vocalizations.

Animals↗