Search PubMed⌕ Search

Biomedical subjects

A J Gabor

Publications and source records attributed to A J Gabor.

At least 19 recordsLinked to original sources
PubMed1998

Seizure detection using a self-organizing neural network: validation and comparison with other detection strategies.

OBJECTIVE: A previously described seizure detection algorithm (CNET) (Gabor, A.J., Leach, R.R. and Dowla, F.U. Automated seizure detection using a self-organizing neural network. Electroenceph. clin. Neurophysiol., 1996, 99: 257-266) was validated with 200 records from 65 patients (4553.8 h of recording) containing 181 seizures. DESIGN AND METHODS: Performance of the algorithm was manifest by its sensitivity ((seizures detected/total seizures) x 100) and selectivity (false-positive errors/Hr-FPH). Comparisons with the Monitor detection algorithm (Version 8.0c, Stellate Systems) and audio-transformation (Oxford Medilog) were performed. RESULTS: CNET detected 92.8% of the seizures and had a mean FPH of 1.35 +/- 1.35. Monitor detected 74.4% of the seizures and had a mean FPH of 3.02 +/- 2.78. Audio-transformation detected all but 3 (98.3%) of the seizures. Selectivity for this detection strategy was not defined. CONCLUSIONS: This study not only validates the CNET algorithm, but also the notion that seizures have frequency-amplitude features that are localized in signal space and can be selectively identified as being distinct from other types of EEG patterns. The ear is a specialized frequency-amplitude detector and when the signal is transformed into audio frequency range (audio-transformation), seizures can be detected with better sensitivity as compared to the other strategies examined.

Adolescent↗
PubMed1996

Automated seizure detection using a self-organizing neural network.

An algorithm for automated seizure detection using the self-organizing map (SOM) neural network (NN), with unsupervised training, was used to detect seizures in 24 long-term EEG recordings. The detection paradigm was tested on a constant 8 channel subset of 18 channel scalp EEG recordings. The NN was trained to recognize seizures using 98 training examples. A strategy was devised using wavelet transform to construct a filter that was 'matched' to the frequency features of examples used to train the NN. Four second epochs of training examples and EEGs being tested were transformed into time-independent representations of spectrograms resulting in a time-frequency representation of the time-series. Rule-based long and short term contextual features were used for detection in association with the NN. Fifty-six seizures were detected from a possible 62 (90%) associated with an average 0.71 +/- 0.79 false-positive errors per hour using the same 'population' detection parameters. When the sensitivity for detection was increased, all but one of the 62 seizures were detected (98%). Less than 1.0 false-positive error per hour occurred in all but 5 records when using the 'population' parameters. The combination of rule-based detection criteria employing contextual parameters and unsupervised training of NNs to recognize time-frequency patterns is a promising direction for automated seizure detection.

Adolescent↗
PubMed1992

Automated interictal EEG spike detection using artificial neural networks.

Feed-forward, error-back-propagation artificial neural networks were applied to recognition of epileptiform patterns in the EEG. The inherent network properties of generalization and variability tolerance were effective in identifying wave forms that differed from the training patterns but still maintained 'epileptiform' spatio-temporal characteristics. The certainty of recognition was measured as a continuous function with a range of 0-1. Two levels of certainty (0.825 and 0.900) were used to indicate recognition of spikes and sharp waves (SSW). An average 94.2% (+/- 7.3) of the SSW were recognized; 20.9% (+/- 22.9) of all recognized SSW were false-positive recognitions. The time required for pattern recognition was well within the time required for digitizing the analogue data. This study provides evidence that neural network technology is, in principle, an effective pattern recognition strategy for identification of epileptiform transients in the EEG. The analysis is sufficiently rapid to be of potential value as a strategy for data reduction of long recordings stored on bulk media.

Electroencephalography↗
PubMed1991

Scleroderma and central nervous system vasculitis.

We describe a patient with scleroderma (CREST syndrome) and central nervous system vasculitis. While angiography demonstrated segmental symmetrical arterial narrowing characteristic of vasculitis, results of leptomeningeal biopsy were normal. There was no evidence of systemic vasculitis, renal failure, or malignant hypertension previously thought to be required to explain central nervous system dysfunction in patients with scleroderma. Signs and symptoms attributable to vasculitis were reversible with aggressive immunosuppressive therapy.

Cerebrovascular Circulation↗
PubMed1990

Electroencephalography laboratory diagnosis of prolonged QT interval.

Patients with prolongation of the QT interval are at risk for significant neurological morbidity and mortality secondary to ventricular tachyarrhythmias. These patients frequently undergo electroencephalographic (EEG) examination to evaluate episodes of loss of consciousness, which may be associated with convulsions. Electrocardiogram recording as a part of the EEG is a simple and common practice, but analysis for possible QT prolongation is not routinely performed by electroencephalographers. This is, in part, due to the fact that while calculation of the corrected QT interval is straight forward, a calculator is generally required. A nomogram that is presented simplifies determination of the corrected QT interval, facilitating diagnosis of prolongation of the QT interval in the EEG laboratory.

Adult↗
PubMed1989

Orientation discrimination sensitivity of single units in cat primary visual cortex.

Responses of visual cortex (area 17) neurons to moving oriented stimuli were recorded from anesthetized cats. The variance of response (SD2) to repeated identical stimuli was directly proportional to response magnitude (R), (SD2 = C2R). The values of C were not found to differ significantly between different types of cortical cells. The relationship predicts that the coefficient of variation (SD/R) will be smallest near the peak of the tuning curve, indicating that the peak response is most reliable for detecting an orientation but not necessarily the most sensitive to a change in orientation. Tuning curves and response variability were then examined to determine the orientation at which the neuron was most sensitive to changes in stimulus orientation using signal detection theory. The discrimination index (d' = [R1-R2]/SD) for a 1 degree change in stimulus orientation was greatest along the flanks of the tuning curve. In order to generalize the experimental data, response distributions derived from a model of cells with parameters based on experimental data were examined to determine the minimal discriminable change in stimulus orientation. Changes of stimulus orientation between 0.6 and 5 deg of arc could be detected from single responses of a single cell by an optimal observer with 75% accuracy if the orientation change was centered at the most sensitive part of the tuning curve.

Action Potentials↗
PubMed1988

Treatment of status epilepticus: a prospective comparison of diazepam and phenytoin versus phenobarbital and optional phenytoin.

In a randomized, nonblinded clinical trial, 36 consecutive patients with generalized convulsive status epilepticus were treated with either combination diazepam and phenytoin (DZ/DPH) or phenobarbital (PB). Phenytoin was added to the PB regimen if seizures persisted for 10 minutes after beginning therapy. The cumulative convulsion time (total time spent in active convulsive movements) was shorter for the PB group than for the DZ/DPH group (median, 5 versus 9 minutes, p less than 0.06); the response latency (elapsed time from initiation of therapy to the end of the last convulsion) was also shorter for the PB group (median, 5.5 versus 15 minutes, p less than 0.10). The median cumulative convulsion time is between 0 and 14 minutes shorter for the PB regimen than for the DZ/DPH regimen (95% confidence interval). Similarly, the median response latency for the PB regimen is between 1 minute longer and 20 minutes shorter than that for the DZ/DPH regimen (95% confidence interval). The frequencies of intubation, hypotension, and arrhythmias were similar in the two groups. Eleven of 18 patients in the PB group responded to phenobarbital monotherapy. We conclude that the PB regimen is rapidly effective, comparable in safety, and enjoys certain practical advantages in comparison with the DZ/DPH regimen.

Adult↗
PubMed1987

Generators of human spinal somatosensory evoked potentials.

Somatosensory evoked potentials recorded over the spine with a noncephalic reference following posterior tibial nerve stimulation have several components. (1) A stationary, synapse-dependent, negative potential (N22) occurs synchronously with a positive potential, P22, recorded ventral to the spinal cord and is localized to the lumbar region overlying the lumbar root entry zone. The N22/P22 complex is attributed to activation of interneurons in the dorsal gray of the lumbar cord. (2) A traveling negative potential with a gradually increasing latency may be recorded from the sacral to the cervical region. Its short refractory period indicates that it is not dependent on transmission across a synapse. This activity is attributed to transmission of the afferent volley through the lumbosacral plexus, roots, and the dorsal columns of the spinal cord. (3) N29, a stationary, synapse-dependent negative potential, localizes to the rostral cervical spine and is attributed to activation of the gracile nucleus relay cells. Following stimulation of the median nerve or fingers, the waveforms recorded over the cervical spine with a noncephalic reference include (1) the proximal plexus volley, a traveling negative potential reflecting transmission through the proximal brachial plexus and roots; (2) the dorsal column volley (DCV), the latency of which gradually increases from the caudal to rostral cervical region (the DCV is attributed to transmission of the afferent volley through the dorsal columns of the cervical cord); and (3) N13, a stationary negative waveform, with a long refractory period consistent with its dependence on transmission across a synapse. Experimental animal and human studies indicate that the N13 waveform is dependent on activity of at least two generator sites, namely the dorsal gray of the cervical cord and the cuneate nucleus.

Evoked Potentials, Somatosensory↗
PubMed1987

Effect of movement on human spinal and subcortical somatosensory evoked potentials.

Sensory transmission in dorsal column nuclei is inhibited during voluntary movement in experimental animals. We have studied the human response by recording spine and scalp somatosensory evoked potentials. Finger movement attenuated the amplitude and duration of the cervical N13 and the scalp N18 and N20 waves. Foot movement did not alter the lumbar N22 after foot stimulation, but the scalp P38 was attenuated. N22 results solely from activation of interneurons in the dorsal gray of the cord at the root entry zone, but N13 may receive contributions from the nucleus cuneatus. Therefore, the movement-induced attenuation of N13 is attributed to decreased contribution from the nucleus cuneatus.

Action Potentials↗
PubMed1987

Cervical synapse-dependent somatosensory evoked potential following posterior tibial nerve stimulation.

We have demonstrated the presence of a localized, synapse-dependent negativity (N29) recorded over the upper cervical spine after bilateral stimulation of the posterior tibial nerves at the ankle. The amplitude of N29 is maximal at the level of the second cervical spine and decreases at more rostral and caudal levels. The peak latency of N29 remains constant at all levels. N29 has a long refractory period when compared with the refractory period of the afferent volley recorded at either the sacral or thoracic level. N29 is most likely generated by activation of the nucleus gracilis by the afferent volley. The cervical N13 after median nerve stimulation probably has multiple generator sites, including the nucleus cuneatus.

Electric Stimulation↗
PubMed1986

Effect of sleep on the electrographic manifestations of epilepsy.

Alterations in the level of arousal have profound predictable effects on the electrographic manifestations of epileptogenic abnormalities. The changes produced by sleep are dramatic in patients with generalized epilepsy as compared to patients with partial epilepsy. The epileptiform complexes associated with the generalized epilepsies show changes in spatial distribution, voltage, temporal sequencing, and waveform morphology. The changes are qualitatively similar in the different types of generalized seizure disorders but vary in the degree to which they are expressed in proportion to the "severity" of the seizure disorder. The presence of a generalized epileptogenic abnormality may result in alteration of K-complex waveform morphology. Multifocal spikes occurring in conjunction with bilateral synchronous discharges are frequently represented by a spatial field distribution consistent with a "horizontal dipole," during sleep especially in children.

Adult↗
PubMed1985

Response covariance in cat visual cortex.

The activity of pairs of neurons in the visual cortex (area 17) of anaesthetized, paralysed cats was recorded using two independently manipulated micropipettes. The number of spikes in the evoked responses of pairs of single neurons were analyzed for response covariance. Responses of the majority of cell pairs (83%) did not covary. Covariance was restricted to closeby neurons with distances of less than 150 microns and with identical orientation and ocular dominance preference.

Analysis of Variance↗
PubMed1985

The human posterior tibial somatosensory evoked potential: synapse dependent and synapse independent spinal components.

Evidence has been obtained for the existence of two separate events occurring in the human spinal cord following posterior tibial nerve (PTN) stimulation. These events can be recorded on the surface in unanesthetized individuals. The first is an ascending wave which is conducted up to the cord at constant velocity and has a relatively short refractory period consistent with a compound nerve action potential. This represents the afferent volley traversing the lumbosacral plexus and the ascending dorsal columns. A second event, the N22/P22 complex, is surface negative on the back and surface positive anteriorly; its amplitude is maximal 5-15 cm above the level of the L4 spine and its peak latency remains constant at all levels. This activity has a relatively long refractory period. These characteristics of N22/P22 indicate that it is a localized synaptically dependent event conforming to a transverse dipole with dorsal negativity and a simultaneous anterior positivity. The N22/P22 is probably generated in the dorsal grey at the root entry zone. The N22/P22 is analogous to the stationary N13/P13 recorded over the neck following median nerve stimulation.

Adult↗
PubMed1984

Intracranial pressure during epileptic seizures.

A comatose 31-year-old male with presumed viral encephalitis and frequent partial motor seizures was paralyzed with pancuronium in an attempt to reduce recurrent elevation of intracranial pressure (ICP) associated with each seizure. ICP was continuously monitored with a Richmond Bolt and 5 electrographic seizures originating in the left frontal area were recorded. Each ictal episode was associated with stable blood pressure and an increase of ICP. The average seizure duration was 78 +/- 17 sec (mean +/- S.D.) and the average maximum increase of ICP above baseline during the seizures was 6.5 +/- 0.6 mm Hg with average peak ICP of 16.0 +/- 0.86 mm Hg. A simple mathematical model predicts the rate of increase of ICP, the peak ICP, the phase difference between maximum spike frequency and maximum ICP, and the rate at which ICP returns to pre-ictal values after termination of the seizure. The predicted values of ICP closely approximate the experimentally derived data. Therefore, the time course of the ICP appears to be determined by the frequency of the fundamental units of abnormal synchronized activity (the epileptogenic spike) and the CSF pressure-volume dynamics existing at the time of the seizure. An average increment of ICP per spike can be calculated for each seizure. The model also predicts that patients may develop high ICPs due to prolonged seizures. Prolonged unrecognized seizures may occur in patients who are therapeutically paralyzed as demonstrated by the case described here.

Adult↗
PubMed1980

Movement-induced seizures in nonketotic hyperglycemia.

We studied two patients with hyperglycemia and focal seizures induced by repetitive movement. Treatment of the hyperglycemia controlled the seizures. Early diagnosis is necessary for institution of appropriate therapy, and to decrease the morbidity associated with nonketotic hyperglycemic coma, which may evolve.

Aged↗