Discriminating between bipolar and major depressive disorder using a machine learning approach and resting-state EEG data.

OBJECTIVE: Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge as effective treatment is quite different for each condition. In this study electroencephalography (EEG) was explored as an objective biomarker for distinguishing MDD from BD using an efficient machine learning algorithm (MLA) trained by a relatively large and balanced dataset. METHODS: A 3 step MLA was applied: (1) a multi-step preprocessing method was used to improve the quality of the EEG signal, (2) symbolic transfer entropy (STE), an effective connectivity measure, was applied to the resultant EEG and (3) the MLA used the extracted STE features to distinguish MDD (N = 71) from BD (N = 71) subjects. RESULTS: 14 connectivity features were selected by the proposed algorithm. Most of the selected features were related to the frontal, parietal, and temporal lobe electrodes. The major involved regions were the Broca region in the frontal lobe and the somatosensory association cortex in the parietal lobe. These regions are near electrodes FC5 and CPz and are involved in processing language and sensory information, respectively. The resulting classifier delivered an evaluation accuracy of 88.5% and a test accuracy of 89.3%, using 80% of the data for training and evaluation and the remaining 20% for testing, respectively. CONCLUSIONS: The high evaluation and test accuracies of our algorithm, derived from a large balanced training sample suggests that this method may hold significant promise as a clinical tool. SIGNIFICANCE: The proposed MLA may provide an inexpensive and readily available tool that clinicians may use to enhance diagnostic accuracy and shorten time to effective treatment.

A Machine Learning Algorithm to Discriminating Between Bipolar and Major Depressive Disorders Based on Resting EEG Data.

Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge due to the lack of known biomarkers. Conventional methods of diagnosis rest exclusively on symptomatic presentation, and personal and family history. As a result, BD-depressed episode (BD-DE) is often misdiagnosed as MDD, and inappropriate therapy is given. Electroencephalography (EEG) has been widely studied as a potential source of biomarkers to differentiate these disorders. Previous attempts using machine learning (ML) methods have delivered insufficient sensitivity and specificity for clinical use, likely as a consequence of the small training set size, and inadequate ML methodology. We hope to overcome these limitations by employing a training dataset of resting-state EEG from 71 MDD and 71 BD patients. We introduce a robust 3 steps ML technique: 1) a multi-step preprocessing method is used to improve the quality of the EEG signal 2) symbolic transfer entropy (STE), which is an effective connectivity measure, is applied to the resultant EEG signals 3) the ML algorithm uses the extracted STE features to distinguish MDD from BD patients. Clinical Relevance--- The accuracy of our algorithm, derived from a large sample of patients, suggests that this method may hold significant promise as a clinical tool. The proposed method delivered total accuracy, sensitivity, and specificity of 84.9%, 83.4%, and 87.1%, respectively.

Menstrual effects on asymmetrical olfactory acuity.

Gender specific discrepancies on psychometric examination are often interpreted to reflect static differences in cerebral hemisphere specialization, but dynamic alterations relating to circulating gonadal hormones may also be relevant after puberty. The often cited inference of a right hemisphere advantage in males and left hemisphere advantage in females derived from small but reliable differences on spatial tasks and verbal tasks, for example, may to some extent relate to gender-specific differences in circulating gonadal hormones. Performance fluctuations on other higher order cognitive tasks through the menstrual cycle tend to support a temporal association between alterations in cerebral laterality and hormone fluctuations. A potential left hemisphere advantage after menstruation when estrogen and progesterone levels are high in contrast to a right hemisphere advantage at menstruation when estrogen and progesterone levels are low has also received support from shifts in visual field perception. The present investigation continues this line of work by measurement of prospective changes in unirhinal olfactory acuity in the menstrual, ovulatory, and midluteal phases of the menstrual cycle in 11 healthy women who agreed to blood assays of estradiol and progesterone prior to completing a modified version of the Connecticut Chemosensory Perception Exam (CCPE). The CCPE detection of n-butanol showed a clear pattern of changes over the menstrual cycle marked by an asymmetry favoring the right nostril during menstruation when estradiol and progesterone levels were low, an asymmetry favoring the left nostril during ovulation when estradiol levels were high and progresterone levels were low, and an absence of asymmetry during the midluteal phase when estradiol levels decreased and progesterone levels increased. Preliminary correlation analyses revealed a potential competitive influence of estradiol and progesterone on this apparent shift in cerebral laterality. There is thus sufficient evidence to conclude that dynamic changes in relative cerebral hemisphere advantages have a temporal relation to fluctuations in circulating gonadal hormones and to suggest the value of additional investigation of more specific causal relations.

Asymmetrical hand force persistence and neuroleptic treatment in schizophrenia.

The recent development of an isometric instrument for the precise quantification of hand force persistence has created a novel opportunity for the evaluation of potential motor asymmetries in schizophrenia and their response to treatment. A study of asymmetries in the unmedicated state may provide insight into the pathogenesis of schizophrenia, whereas alterations of asymmetries in response to antipsychotic medication could assist the delineation of a cerebral mechanism for the effects of pharmacotherapy. The hand force persistence of 21 unmedicated patients with schizophrenia was compared to 21 age, gender, and handedness matched normal controls. The effect of neuroleptic treatment on hand force persistence was then evaluated on a subset of 10 patients after at least 30 days of treatment. The anticipated asymmetry was evident in the unmedicated sample that showed impaired right hand force persistence compared to the normal control sample. The prospective comparison showed an alleviation of the asymmetry resulting from an improvement of right hand force persistence with treatment. In addition to providing further support to a primary left hemisphere cerebral involvement in schizophrenia, the present results suggest that prior investigations of motor asymmetry may have been compromised by the study of medicated patients. The apparently paradoxical improvement of motor skill may relate to the substantial number of patients treated with 2nd generation neuroleptic medications which may implicate an improvement in left hemisphere physiology in the cognitive advantages of the novel treatments.

Low-resolution electrical tomography of the brain during psychometrically matched verbal and spatial cognitive tasks.

EEGs were recorded from 75 normal, young, female subjects during psychometrically matched verbal (WF) and spatial (DL) cognitive tasks to elicit the differences in the electrical source distribution inside the brain. Recordings were obtained using 43 EEG and 3 guard electrodes then visually edited and spatially filtered to remove extracerebral artifacts. Twenty 1-sec artifact-free epochs were obtained and analyzed from 42 and 60 subjects during WF and DL respectively. Of these subjects, 20 were placed in a training set and the remainder into a test set. The baseline for the comparison of the two tasks was established by factoring the average cross-spectral matrices of the training-set EEGs, computed in the theta, alpha, and beta frequency bands into spatial patterns common to the two tasks. Only those spatial patterns that contributed to the correct classification of subjects in the test set were included in the source analysis. The source-current density distributions were obtained using the LORETA-KEY algorithm. The results show that the source-current density distribution is related to the putative functional activity in the brain in all three frequency bands. The electrical effects of the tasks are both most highly localized and lateralized in the theta band. The effects in the alpha and beta bands are much more generalized and are strongly lateralized only during one and the other of the tasks respectively. The conclusion is that WF is mainly a left central and bilateral frontal cerebral process while DL is mainly a right central and bilateral posterior cerebral process.

Asymmetrical olfactory acuity and neuroleptic treatment in schizophrenia.

Uni-rhinal olfactory acuity in schizophrenia was investigated in two experiments. The first assessed the presence of a predicted atypical asymmetry of nostril laterality and the second assessed the effect of antipsychotic treatment on the asymmetry. Although olfactory identification impairment has been well documented in schizophrenia, olfactory acuity has been neglected. This may be an oversight as cerebral structures of the mesial temporal lobe important to olfactory perception have often been implicated in the pathophysiology of schizophrenia and it is thus reasonable to postulate a primary impairment of olfactory acuity in schizophrenia. In addition, unmedicated patients with schizophrenia have exhibited asymmetrical laterality favouring the right over the left hemisphere in studies of visual, haptic, and auditory perception, and the few published prospective treatment studies have suggested a reversal of this asymmetry with first generation neuroleptic treatments. In experiment 1 a generalization of the perceptual asymmetry to olfactory acuity was examined by measurement of n-butanol olfactory thresholds with the Connecticut Chemosensory Perception Exam (CCPE) in an unmedicated sample of 17 patients with schizophrenia and 17 age, gender, and handedness matched normal controls. The patient sample showed an asymmetrical impairment of the left nostril that was not apparent in the normal control sample. In experiment 2, the CCPE was administered to a new sample of 10 patients with schizophrenia before and after neuroleptic treatment. The asymmetry observed in experiment 1 was replicated, and the relative advantage of the right nostril shifted to a relative advantage of the left nostril over the course of 8weeks of treatment. Results are discussed in relation to cerebral aspects of schizophrenia and potential implications to cognitive change from treatment.

Discriminant analysis and equivalent source localization of the EEG related to cognitive functions.

Discriminant analysis and EEG source localization methods were employed to compare groups of normal subjects during different cognitive conditions using 43-channel EEG recordings in the alpha (8-13 Hz) frequency band. Recordings were obtained from 69 dextral females during 2 passive conditions, Eyes-Open and Eyes-Closed, and 2 active conditions, Word-Finding and Dot-Localization. The cross-spectral matrix between all of the electrode sites was used to characterize the EEGs obtained during each condition. The subjects were partitioned into training and test sets and quadratic discriminant functions were constructed from the training sets to classify the EEGs. The discriminant functions successfully classified both the training and test sets at rates approaching 80%. The classification was repeated using only the diagonal (power spectral) elements of the cross-spectral matrices in the discriminant functions and this approach was successful in discriminating between the EEGs from the passive cognitive conditions but failed to discriminate between the EEGs from the active conditions. Source localization using a modified MUSIC algorithm indicated that the centers of brain electrical activity that distinguished the Eyes-Closed condition from the Eyes-Open condition were located in the medial occipital and right frontal regions. Centers of electrical activity that distinguished the Word-Finding condition from the Dot-Localization condition were located in the right medial posterior and left temporal regions. Validation of the locations of the centers of activity was accomplished by repeating the classification procedures using the spatial patterns generated on the scalp by dipole current sources placed at these locations.

Frequency domain discriminant analysis of the electroencephalogram.

A frequency domain generalization of the classical quadratic discriminant function was applied to the problem of classifying alpha-band multichannel electroencephalogram recordings in three task conditions. The data consisted of 41-channel recordings obtained in eyes closed, verbal, and spatial task conditions. Classifier performance was measured by deriving a decision rule from a training sample of 42 recordings and then applying the obtained rule to a test sample of 46 recordings. The proportion of correct classification was .93 in the training sample and .85 in the test sample. The classifier performed better when based on the complete cross-spectral matrix than when restricted to power spectrum variables. Classification based on a subset of 16 leads reduced the overall proportion of correct classification to .79 in the training sample and to .70 in the test sample.

Spatial patterns in the background EEG underlying mental disease in man.

The spatial patterns underlying differences in the background EEGs of schizophrenic, manic and depressed patients and a group of normal controls has been examined during the eyes open and eyes closed resting conditions and during 3 cognitive tasks. The method of principal-component analysis was used to extract spatial patterns which are common to the EEGs of 2 groups but which account for maximally different proportions of the combined variances. The common spatial patterns in all possible pairings of the groups were used to extract variance-related feature vectors from the individual EEG epochs in the 2 groups and the means of these vectors were subjected to statistical analyses. The results of these analyses indicate that there are significant differences in the EEGs from all 4 of the groups. The spatial patterns underlying the features which are significantly different in each comparison are shown graphically and used to suggest which brain regions might be implicated in each of the psychiatric conditions and how these are affected by the cognitive condition. The main results are that the EEGs in the schizophrenic group can be characterized by left-sided hyperactivity, in the depressed group by right-sided hyperactivity and in the manic group by bilateral hyperactivity and that these characteristics are best elicited by different cognitive states.

Quantitative EEG studies of pedophilia.

Ninety six pedophiles, whose sexual orientation was confirmed by phallometric response to sexual stimuli, were investigated with quantitative EEG and compared to age- and sex-matched healthy controls. The EEG analysis showed a pattern of increased frontal delta, theta and alpha power (especially during verbal processing) and a pattern of reduced interhemispheric and increased intrahemispheric-interhemispheric coherence, right and left (only during verbal processing), an effect that was restricted to those who showed maximal erotic arousal for sexual partners aged 6-12 years. These findings will be discussed in the context of recent studies which suggest that sexual deviations in the male relate to altered dominant hemispheric functions with disruption of frontal interhemispheric relationships.

Neurophysiological and neuropsychological study of two cases of multiple personality syndrome and comparison with chronic hysteria.

Two cases of multiple personality were studied neurophysiologically and neuropsychologically. Bilateral frontal (Right greater than Left) and left temporal dysfunction was present in both cases, on neuropsychological indicators. Both cases on EEG analysis, were in a state of relative left hemisphere activation, across all cerebral regions and task conditions. The one case who was cured with hypnotherapy, after recovery showed normal left hemisphere functions neuropsychologically but remained in a state of relative left hemisphere activation electrophysiologically. This is in contrast to women with chronic hysteria who exhibit relative right hemisphere activation in all regions and across all conditions. Both patients were unmedicated throughout. A neurophysiological model to account for these findings is presented.

[The obsessive-compulsive syndrome: reflection of fronto-caudate dysregulation of the left hemisphere?]. / Le syndrome obsessionnel-compulsif: reflet d'un défaut de régulation fronto-caudée de l'hémisphère gauche?

Neuropsychological studies of the obsessional syndrome reveal bilateral frontal dysfunction. Early onset is characteristic of males, who have a more chronic course than females, and who are more often non-dextral. Hypermetabolic activity of the frontal and caudate, bilaterally, is found in most PET investigations, although one report is of relative hypofrontality and global cortical hypometabolism for absolute values. Given the frequency with which basal ganglia disease leads to obsessive-compulsive phenomena (Cf Sydenham's Chorea, Gilles de La Tourette) and frontal lobe inertia, a perturbation of frontal-caudate regulatory motor and ideational sub-systems in the obsessional syndrome appears probable. Bilateral caudate atrophy on CT scan has been reported. Further, EP investigations, both somatosensory and auditory, implicate the left hemisphere in obsessions; together with reduced P300 latencies and during imaginal flooding there is increased left (frontal) hemisphere flow (rCBF). Single case studies document the relationship of obsessions to left frontal-left caudate unilateral pathology. It is suggested that lateralized dysregulation of the left fronto-caudate network is the major cerebral determinant of obsessive-compulsive states.

Neuropsychology and psychopathology: a progress report.

This paper summarizes and discusses the contributions of neuropsychological assessment to various forms of psychopathology. Emphasis is placed upon studies done with the Halstead-Reitan battery and the Wechsler Adult Intelligence Scale, but studies done with other neuropsychological test procedures are also reviewed. The conclusions reached are that neuropsychological tests are sensitive to functional regional brain disorganization in psychopathology, and that they are useful in the diagnostic process for a number of disorders including schizophrenia, psychopathy, mood disorders, and other psychiatric conditions.

Influence of gender in schizophrenia as related to other psychopathological syndromes.

The evidence indicating that the forms of schizophrenia in men and women represent different morbid states is reviewed. Age of onset and gender are considered to be of fundamental importance in determining the different symptomatological and evolutionary features of the syndrome in the two sexes. Early-onset forms in males are associated with chronicity, absence of familial predisposition for psychosis, and the presence of structural cerebral pathology specifically involving the dominant hemisphere. Later onset forms in females are characterized by more florid symptoms, more affective features, more familial psychosis, and more favorable outcome with no or less pronounced structural cerebral involvement. It is argued that these differential characteristics derive from the differential hemispheric organization of the male and female brain--which also determines the male susceptibility to other psychopathological syndromes such as psychopathy and sexual deviations as well as the excess in women of schizoaffective states, affective disorders, and late-onset schizophrenia.

Neuroleptics reverse attention asymmetries in schizophrenic patients.

Discussions of hemispheric asymmetry in psychopathology are often confounded by the effects of medication. We examined the effect of neuroleptic drugs on attention asymmetries in acutely psychotic patients admitted for the first time to a psychiatric hospital before the initiation of drug treatment and again after a period of treatment with neuroleptics. Overall performance did not change significantly; however, attention asymmetry was clearly related to the medication status of the patient: unmedicated patients showed inattention to the right hemispace, which changed to more prominent left-sided inattention when medicated. A longer time on medication or a higher daily dose were associated with a shift of inattention from the right to left hemispace. This suggests that neuroleptics may normalize left hemisphere performance, at the expense of deteriorated right hemisphere performance.

[Cerebral mechanisms of the efficacy of electroshock therapy]. / Mécanismes cérébraux de l'efficacité des électrochocs.

There are correlations between schizophrenia and the limbic seizure system on the one hand and the manic-depressive or bipolar syndromes and the generalized seizure system on the other hand, which are theoretically related to the different (although overlapping) neural substrates underlying the two major syndromes of psychosis. Evidence is reviewed that indicates that in ECT-responsive depression (with both bilateral and unilateral nondominant ECT) the modus operandi hinges on right-hemispheric neural events. At the same time the relevance of the complex interactions existing between limbic and generalized seizures, REM suppression, right limbic epilepsy and REM activation is discussed as well as the role of carbamazepine in these interactions.

The effect of brain function on coherence patterns in the bipolar EEG.

Studies of the functional organization of the brain based on measurements of coherence in the EEG have, in the past, suffered from a methodological defect which has made interpretation of the results difficult. The effect involved is the use of an active common reference for the recording of the EEG. As a consequence, inferences related to the functional connectivity of brain between the non-referential sites using coherence have probably been wrong. To avoid the problem with the common reference, we have used bipolar derivations of the EEG and used measurements of coherence to reflect synchrony not between individual sites but between regions of the brain. The EEGs in a population of normal volunteers were examined with respect to coherence. Changes in the patterning of coherence were induced by utilizing EEGs from the volunteers during 3 different functional brain states. The first of these was the resting state, the second the verbal motor state and the third, the spatial motor state. The stepwise discriminant analysis method was used to study differences in the patterning of coherences in the 3 states. The results show that the spatial motor state was the most distinct in this regard amongst the 3 states. They results also interpreted as indicating that changes in the patterning of coherence from that in the resting state consisted of both functionally specific and functionally non-specific components.(ABSTRACT TRUNCATED AT 250 WORDS)