Automatic detection of auditory brainstem responses using feature vectors.

The efficiency of several methods for the automatic detection of near threshold auditory brainstem responses (ABRs), based on feature vectors, was assessed on a sample of 374 recordings obtained from neonates. Features are quantitative descriptors of different aspects of the response commonly taken into consideration by expert ABR judges, such as: similarity with a template, reliability of the response, and intrinsic morphological characteristics. The classification methods used were: comparisons of individual features with a threshold, linear discriminant functions based on combination of features, artificial neural networks (ANNs) based on the raw ABR, and ANNs based on combinations of features. The network used was a modified version of the Fuzzy ART Map model. The accuracy of the classification into normal and abnormal (as previously scored by a panel of experts) was assessed with methods from signal detection theory. Our results showed that the approaches based on feature vectors (discriminant function and ANN) performed more efficiently than the ANN with raw data, or the individual features.

[1]  J. C. Principe,et al.  Performance and training strategies in feedforward neural networks: an application to sleep scoring , 1989, International 1989 Joint Conference on Neural Networks.

[2]  M A Bobes,et al.  Comparison of auditory-evoked potential detection methods using signal detection theory. , 1987, Audiology : official organ of the International Society of Audiology.

[3]  J. Victor,et al.  A new statistic for steady-state evoked potentials. , 1991, Electroencephalography and clinical neurophysiology.

[4]  I. Bankman,et al.  Feature-based detection of the K-complex wave in the human electroencephalogram using neural networks , 1992, IEEE Transactions on Biomedical Engineering.

[5]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[6]  J. B. Sidowski,et al.  Experimental methods and instrumentation in psychology , 1966 .

[7]  H. Schimmel.  The (�) Reference: Accuracy of Estimated Mean Components in Average Response Studies , 1967, Science.

[8]  Özcan Özdamar,et al.  Auditory brainstem evoked potential classification for threshold detection by neural networks. II. Effects of input coding, training set size and composition and network size on performance , 1992 .

[9]  S. Grossberg,et al.  A self-organizing neural network for supervised learning, recognition, and prediction , 1992, IEEE Communications Magazine.

[10]  Terence W. Picton,et al.  Aspects of Averaging , 1983 .

[11]  B.H. Jansen.  Artificial neural nets for K-complex detection , 1990, IEEE Engineering in Medicine and Biology Magazine.

[12]  T Grönfors,et al.  Identification of auditory brainstem responses. , 1993, International journal of bio-medical computing.

[13]  P. Valdés-Sosa,et al.  Detection of event related potentials. , 1989, The International journal of neuroscience.

[14]  R. M. Siegel,et al.  Pattern recognition of the electroencephalogram by artificial neural networks. , 1993, Electroencephalography and clinical neurophysiology.

[15]  R A Dobie,et al.  Objective response detection in the frequency domain. , 1993, Electroencephalography and clinical neurophysiology.

[16]  D. T. Freeman.  Computer Applications in Otolaryngology: Computer Recognition of Brain Stem Auditory Evoked Potential Wave V by a Neural Network , 1992, Proceedings. Symposium on Computer Applications in Medical Care.

[17]  M E Jernigan,et al.  Classification of brain-stem auditory evoked potentials by syntactic methods. , 1986, Electroencephalography and clinical neurophysiology.

[18]  Rafael E. Delgado,et al.  SYNTACTIC RECOGNITION OF AUDITORY BRAINSTEM EVOKED POTENTIALS FOR NEUROLOGICAL DIAGNOSIS. , 1987 .

[19]  T. Picton,et al.  Prognostic validity of brainstem electric response audiometry in infants of a neonatal intensive care unit. , 1991, Audiology : official organ of the International Society of Audiology.

[20]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[21]  C Elberling,et al.  Objective detection of averaged auditory brainstem responses. , 1984, Scandinavian audiology.

[22]  James P. Egan,et al.  PSYCHOPHYSICS AND SIGNAL DETECTION. , 1962 .

[23]  M. Hyde,et al.  Audiometric accuracy of the click ABR in infants at risk for hearing loss. , 1990, Journal of the American Academy of Audiology.

[24]  Özcan Özdamar,et al.  Auditory brainstem evoked potential classification for threshold detection by neural networks. I. Network design, similarities between human-expert and network classification, feasibility , 1992 .

[25]  R. Coppola,et al.  Signal to noise ratio and response variability measurements in single trial evoked potentials. , 1978, Electroencephalography and clinical neurophysiology.

[26]  Terence W. Picton,et al.  Clinical usefulness of auditory evoked potentials: A critical evaluation. , 1991 .

[27]  G. P. Madhavan,et al.  Use of attributed grammars for pattern recognition of evoked potentials , 1988, IEEE Trans. Syst. Man Cybern..

[28]  R. Bickford,et al.  Brain stem auditory evoked potentials: the use of noise estimate. , 1980, Electroencephalography and clinical neurophysiology.

[29]  Jacobson Jt,et al.  Automated and conventional ABR screening techniques in high-risk infants. , 1990 .

[30]  M. Sakai,et al.  Cross-correlation function in the analysis of auditory brainstem response in spinocerebellar degeneration. , 1991, Audiology : official organ of the International Society of Audiology.

[31]  F Y Wu,et al.  A neural network design for event-related potential diagnosis. , 1993, Computers in biology and medicine.

[32]  J A Swets,et al.  Measuring the accuracy of diagnostic systems. , 1988, Science.