Diffusion imaging markers of bipolar versus general psychopathology risk in youth at-risk.

Bipolar disorder (BD) is highly heritable. Thus, studies in first-degree relatives of individuals with BD could lead to the discovery of objective risk markers of BD. Abnormalities in white matter structure reported in at-risk individuals could play an important role in the pathophysiology of BD. Due to the lack of studies with other at-risk offspring, however, it remains unclear whether such abnormalities reflect BD-specific or generic risk markers for future psychopathology. Using a tract-profile approach, we examined 18 major white matter tracts in 38 offspring of BD parents, 36 offspring of comparison parents with non-BD psychopathology (depression, attention-deficit/hyperactivity disorder), and 41 offspring of healthy parents. Both at-risk groups showed significantly lower fractional anisotropy (FA) in left-sided tracts (cingulum, inferior longitudinal fasciculus, forceps minor), and significantly greater FA in right-sided tracts (uncinate fasciculus and inferior longitudinal fasciculus), relative to offspring of healthy parents (P < 0.05). These abnormalities were present in both healthy and affected youth in at-risk groups. Only offspring (particularly healthy offspring) of BD parents showed lower FA in the right superior longitudinal fasciculus relative to healthy offspring of healthy parents (P < 0.05). We show, for the first time, important similarities, and some differences, in white matter structure between offspring of BD and offspring of non-BD parents. Findings suggest that lower left-sided and higher right-sided FA in tracts important for emotional regulation may represent markers of risk for general, rather than BD-specific, psychopathology. Lower FA in the right superior longitudinal fasciculus may protect against development of BD in offspring of BD parents.

Reward-related neural activity and structure predict future substance use in dysregulated youth.

BACKGROUND: Identifying youth who may engage in future substance use could facilitate early identification of substance use disorder vulnerability. We aimed to identify biomarkers that predicted future substance use in psychiatrically un-well youth. METHOD: LASSO regression for variable selection was used to predict substance use 24.3 months after neuroimaging assessment in 73 behaviorally and emotionally dysregulated youth aged 13.9 (s.d. = 2.0) years, 30 female, from three clinical sites in the Longitudinal Assessment of Manic Symptoms (LAMS) study. Predictor variables included neural activity during a reward task, cortical thickness, and clinical and demographic variables. RESULTS: Future substance use was associated with higher left middle prefrontal cortex activity, lower left ventral anterior insula activity, thicker caudal anterior cingulate cortex, higher depression and lower mania scores, not using antipsychotic medication, more parental stress, older age. This combination of variables explained 60.4% of the variance in future substance use, and accurately classified 83.6%. CONCLUSIONS: These variables explained a large proportion of the variance, were useful classifiers of future substance use, and showed the value of combining multiple domains to provide a comprehensive understanding of substance use development. This may be a step toward identifying neural measures that can identify future substance use disorder risk, and act as targets for therapeutic interventions.

Predicting clinical outcome from reward circuitry function and white matter structure in behaviorally and emotionally dysregulated youth.

Behavioral and emotional dysregulation in childhood may be understood as prodromal to adult psychopathology. Additionally, there is a critical need to identify biomarkers reflecting underlying neuropathological processes that predict clinical/behavioral outcomes in youth. We aimed to identify such biomarkers in youth with behavioral and emotional dysregulation in the Longitudinal Assessment of Manic Symptoms (LAMS) study. We examined neuroimaging measures of function and white matter in the whole brain using 80 youth aged 14.0 (s.d.=2.0) from three clinical sites. Linear regression using the LASSO (Least Absolute Shrinkage and Selection Operator) method for variable selection was used to predict severity of future behavioral and emotional dysregulation measured by the Parent General Behavior Inventory-10 Item Mania Scale (PGBI-10M)) at a mean of 14.2 months follow-up after neuroimaging assessment. Neuroimaging measures, together with near-scan PGBI-10M, a score of manic behaviors, depressive behaviors and sex, explained 28% of the variance in follow-up PGBI-10M. Neuroimaging measures alone, after accounting for other identified predictors, explained ~1/3 of the explained variance, in follow-up PGBI-10M. Specifically, greater bilateral cingulum length predicted lower PGBI-10M at follow-up. Greater functional connectivity in parietal-subcortical reward circuitry predicted greater PGBI-10M at follow-up. For the first time, data suggest that multimodal neuroimaging measures of underlying neuropathologic processes account for over a third of the explained variance in clinical outcome in a large sample of behaviorally and emotionally dysregulated youth. This may be an important first step toward identifying neurobiological measures with the potential to act as novel targets for early detection and future therapeutic interventions.

Altered functioning of reward circuitry in youth offspring of parents with bipolar disorder.

BACKGROUND: Offspring of parents with bipolar disorder (BD) (BO) are at higher risk of BD than offspring of parents with non-BD psychopathology (NBO), although both groups are at higher risk than offspring of psychiatrically healthy parents (HC) for other affective and psychiatric disorders. Abnormal functioning in reward circuitry has been demonstrated previously in individuals with BD. We aimed to determine whether activation and functional connectivity in this circuitry during risky decision-making differentiated BO, NBO and HC. METHOD: BO (n = 29; mean age = 13.8 years; 14 female), NBO (n = 28; mean age = 13.9 years; 12 female) and HC (n = 23; mean age = 13.7 years; 11 female) were scanned while performing a number-guessing reward task. Of the participants, 11 BO and 12 NBO had current non-BD psychopathology; five BO and four NBO were taking psychotropic medications. RESULTS: A 3 (group) × 2 (conditions: win-control/loss-control) analysis of variance revealed a main effect of group on right frontal pole activation: BO showed significantly greater activation than HC. There was a significant main effect of group on functional connectivity between the bilateral ventral striatum and the right ventrolateral prefrontal cortex (Z > 3.09, cluster-p < 0.05): BO showed significantly greater negative functional connectivity than other participants. These between-group differences remained after removing youth with psychiatric disorders and psychotropic medications from analyses. CONCLUSIONS: This is the first study to demonstrate that reward circuitry activation and functional connectivity distinguish BO from NBO and HC. The fact that the pattern of findings remained when comparing healthy BO v. healthy NBO v. HC suggests that these neuroimaging measures may represent trait-level neurobiological markers conferring either risk for, or protection against, BD in youth.

The influence of comorbid disorders on the episodicity of bipolar disorder in youth.

OBJECTIVE: Bipolar disorder (BP) frequently co-occurs with other psychiatric disorders. We examine whether course of anxiety disorders (ANX), attention deficit hyperactivity disorder (ADHD), disruptive behavior disorders (DBD), and substance use disorders (SUD) influence likelihood of recovery and recurrence of depression and mania in BP youth. METHOD: Weekly ratings of psychiatric disorder intensity were obtained from 413 participants of the Course and Outcome of BP Youth project, followed for an average of 7.75 years. Multiple-event Cox proportional hazards regression analyses examined worsening of comorbid disorders as predictors of mood episode recovery and recurrence. RESULTS: Increased severity in ANX and SUD predicted longer time to recovery and less time to next depressive episode, and less time to next manic episode. Multivariate models with ANX and SUD found that significant effects of ANX remained, but SUD only predicted longer time to depression recovery. Increased severity of ADHD and DBD predicted shorter time to recurrence for depressive and manic episodes. CONCLUSION: There are significant time-varying relationships between the course of comorbid disorders and episodicity of depression and mania in BP youth. Worsening of comorbid conditions may present as a precursor to mood episode recurrence or warn of mood episode protraction.

Behavioral and emotional dysregulation trajectories marked by prefrontal-amygdala function in symptomatic youth.

BACKGROUND: Neuroimaging measures of behavioral and emotional dysregulation can yield biomarkers denoting developmental trajectories of psychiatric pathology in youth. We aimed to identify functional abnormalities in emotion regulation (ER) neural circuitry associated with different behavioral and emotional dysregulation trajectories using latent class growth analysis (LCGA) and neuroimaging. METHOD: A total of 61 youth (9-17 years) from the Longitudinal Assessment of Manic Symptoms study, and 24 healthy control youth, completed an emotional face n-back ER task during scanning. LCGA was performed on 12 biannual reports completed over 5 years of the Parent General Behavior Inventory 10-Item Mania Scale (PGBI-10M), a parental report of the child's difficulty regulating positive mood and energy. RESULTS: There were two latent classes of PGBI-10M trajectories: high and decreasing (HighD; n=22) and low and decreasing (LowD; n=39) course of behavioral and emotional dysregulation over the 12 time points. Task performance was >89% in all youth, but more accurate in healthy controls and LowD versus HighD (p<0.001). During ER, LowD had greater activity than HighD and healthy controls in the dorsolateral prefrontal cortex, a key ER region, and greater functional connectivity than HighD between the amygdala and ventrolateral prefrontal cortex (p's<0.001, corrected). CONCLUSIONS: Patterns of function in lateral prefrontal cortical-amygdala circuitry in youth denote the severity of the developmental trajectory of behavioral and emotional dysregulation over time, and may be biological targets to guide differential treatment and novel treatment development for different levels of behavioral and emotional dysregulation in youth.

Amygdala response to fearful faces in anxious and depressed children.

BACKGROUND: Alterations in amygdala function have been implicated in the pathophysiological characteristics of adult anxiety and depressive disorders. Studies with healthy adults and children, as well as with adults who have amygdala lesions, have found facial expressions of emotion to be useful probes of amygdala activity. Our study examined the amygdala response to fearful and neutral facial expressions in healthy, anxious, and depressed children. We hypothesized that children with anxiety and depression may show atypical amygdala responses to emotional stimuli. METHODS: Twelve children (8-16 years of age) with generalized anxiety or panic disorder and 12 healthy comparison children underwent noninvasive functional magnetic resonance imaging while viewing photographs of fearful and neutral facial expressions. In a second comparison, 5 girls with major depressive disorder were compared with 5 anxious and 5 healthy girls from the previous sample. RESULTS: Children with anxiety disorders showed an exaggerated amygdala response to fearful faces compared with healthy children, whereas depressed children showed a blunted amygdala response to these faces. In addition, the magnitude of the amygdala's signal change between fearful and neutral faces was positively correlated with the severity of everyday anxiety symptoms. CONCLUSIONS: Our results suggest that amygdala function is affected in both anxiety and depression during childhood and adolescence. Moreover, this disruption appears to be specific to the child's own rating of everyday anxiety.

Relation between anxiety and depressive disorders in childhood and adolescence.

Current research indicates that there is a strong relationship between pediatric anxiety disorders and depression. Assessment measures show high rates of correlation between depression and anxiety and much of the overlap may be related to a common domain of negative affectivity. Anxious youth and depressed youth share a cognitive style marked by a negative bias in information processing. Anxiety disorders and depression are frequently comorbid in children and adolescents. About 25-50% of depressed youth have comorbid anxiety disorders and about 10-15% of anxious youth have depression. Twin and family studies have demonstrated that pediatric anxiety disorders and depression likely share some common genetic factors or influences. Selective serotonin reuptake inhibitors and cognitive-behavioral therapy have been shown in randomized controlled trials to be efficacious for both pediatric depression and anxiety disorders. Integrating the treatment literature with studies of phenomenology, biology and genetics indicates that pediatric anxiety disorders and depression may share a genetically determined neurobiological component that could involve neural circuits that include or are modulated by serotonergic neurons. This component could contribute to the negative affective temperament that appears to be common in both pediatric depression and anxiety disorders.

Classification of in vivo 1H MR spectra from breast tissue using artificial neural networks.

BACKGROUND: The study was designed in order to investigate whether artificial neural networks could be used for analysis of in vivo magnetic resonance (MR) spectra from breast cancer patients. MATERIALS AND METHODS: In vivo 1H MR spectra with three different echo times (TE 135, 350 and 450 msec) were acquired from patients with benign and malignant breast lesions and from healthy volunteers, of whom some were breast-feeding. A spectral region (4.0-1.5 ppm) was used as input for artificial neural network analysis, for the attempted classification of the data into different groups. RESULTS: Data recorded at all three echo times were necessary to obtain the best results. Furthermore, malignant tissue was differentiated from benign tumours using this approach, whereas benign tumours were poorly separated from healthy tissue. CONCLUSION: The results presented here indicate that in vivo MR spectroscopy in conjunction with neural network analysis might be useful for the evaluation of breast lesions.

Feature extraction and classification of dynamic contrast-enhanced T2*-weighted breast image data.

UNLABELLED: The relatively low specificity of dynamic contrast-enhanced T1-weighted magnetic resonance imaging (MR) imaging of breast cancer has lead several groups to investigate different approaches to data acquisition, one of them being the use of rapid T2*-weighted imaging. Analyses of such data are difficult due to susceptibility artifacts and breathing motion. MATERIALS AND METHODS: One-hundred-twenty-seven patients with breast tumors underwent MR examination with rapid, single-slice T2*-weighted imaging of the tumor. Different methods for classifying the image data set using leave-one-out cross validation were tested. Furthermore, a semi-automatic region of interest (ROI) definition tool was presented and compared with manual ROI definitions from a previous study. Finally, pixel-by-pixel analysis was done and compared with ROI analysis. The analyses were done with and without noise reduction. RESULTS: The minimum enhancement parameter was the most robust and accurate of the parameters tested. The semi-automatic ROI definition method was fast and produced similar results as the manually defined ROIs. Noise reduction improved both sensitivity and specificity, but the improvement was not statistically significant. The pixel-based analysis methods used in the present study did not improve classification results. CONCLUSIONS: Analysis of T2*-weighted breast images can be done in a rapid and robust manner by using semi-automatic ROI definition tools in combination with noise reduction. Minimum enhancement gives an indication of malignancy in T2*-weighted imaging.

Low growth hormone response to growth hormone-releasing hormone in child depression.

BACKGROUND: This study examined growth hormone (GH) response to growth hormone-releasing hormone (GHRH) in a large sample of depressed children compared with normal control children. Within-subject comparisons were also performed in control subjects to examine test-retest reliability and in depressed children comparing episode versus clinical recovery. METHODS: The sample included depressed children (n = 82) and normal control children (n = 55) group-matched for age, gender, and pubertal status; the mean ages were 11.2 +/- 1.7 and 11.2 +/- 1.8 years, respectively. We gave GHRH (0.1 mcg/Kg) at 9 AM, and serum GH levels were determined every 15 min from -30 min through +90 min of the GHRH infusion. A subgroup of normal control subjects (n = 11) repeated the protocol for test-retest reliability within a 2-month interval. A subgroup of depressed children (n = 20) were restudied off all medications following full clinical remission from depression. RESULTS: The mean GH response to GHRH was significantly lower in the depressed group (8.7 ng/mL +/- SEM 0.9) compared with normal control children [12.2 ng/mL +/- SEM 1.3; t(135) = 2.59, p =.01 effect size 0.44]. The test-retest reliability of GH response to GHRH was stable (intraclass correlation =.93 for mean post-GH). The GH response to GHRH remained low in subjects restudied during clinical remission from depression. CONCLUSIONS: Depressed children show low GH response to GHRH. The measure appears to be reliable, and the low GH response continues following clinical remission. Further studies are needed to explore the mechanism and relative specificity of this finding.

Anxiety disorders in suicidal youth.

OBJECTIVE: To test the hypothesis that anxiety disorders are associated with suicidal ideation and suicide attempts in a child and adolescent sample referred to a clinic. METHOD: The sample comprised 1979 patients aged 5 to 19 years who were assessed using the Schedule for Affective Disorders and Schizophrenia for School Aged Children--Present Episode (K-SADS-P) at an outpatient mood and anxiety disorders clinic. Subjects were stratified by age and categorized into mutually exclusive groups as being nonsuicidal (n = 817), having suicidal ideation (n = 768), or having attempted suicide (n = 394) in the current episode. Psychiatric diagnoses based on DSM-IV criteria were compared to determine if anxiety disorders were a risk factor for suicidal ideation and suicide attempts. Logistic regression was used to control for significant demographic characteristics and comorbid disorders. RESULTS: After stratifying by age, we found no differences across the 3 groups (ideators, attempters, and nonsuicidal youth) in rates of an anxiety disorder in general or in specific rates of panic disorder, agoraphobia, social phobia, simple phobia, and obsessive-compulsive disorder. Two salient findings involving anxiety disorders were noted with regression analysis. In younger children (age < or = 15 years), attempters had a significantly lower prevalence of separation anxiety disorder (SAD), compared with ideators (OR = 0.30; 95%CI, 0.11 to 0.80; P = 0.006) and nonsuicidal youngsters (OR = 0.14; 95%CI, 0.05 to 0.39; P < 0.0001). In older children (age > 15 years), generalized anxiety disorder (GAD) was more prevalent in ideators (OR = 1.65; 95%CI, 1.03 to 2.66; P = 0.03) than in nonsuicidal patients. CONCLUSIONS: Based on this clinical sample, the relation between pediatric anxiety disorder and suicidal ideation and suicide attempts is not straightforward. However, further studies in nonreferred samples are warranted.

Growth hormone secretion in children and adolescents at high risk for major depressive disorder.

BACKGROUND: Decreased growth hormone (GH) response to pharmacologic stimulation has been found in children and adolescents during an episode of major depressive disorder and after recovery. In this study, we sought to determine whether GH secretion is similarly altered in children and adolescents who had never experienced depression but were at high risk of developing depression. METHODS: Subjects were 8 through 16 years of age and selected for high- and low-risk status according to familial loading for mood disorders. Sixty-four high-risk and 55 low-risk healthy subjects participated in the study, which assessed the following GH measures: (1) GH before growth hormone-releasing hormone (GHRH) infusion, every 15 minutes for 30 minutes; (2) GH response after intravenous infusion of GHRH (0.1 microg/kg), every 15 minutes for 90 minutes; and (3) nocturnal GH every 20 minutes from 9 PM until morning awakening. RESULTS: After stimulation with GHRH, the high-risk subjects secreted significantly less GH compared with the low-risk healthy controls (effect sizes for mean and peak GH, 0.52 [P =.007] and 0.40 [P =.04], respectively). In contrast, there were no between-group differences in the pre-GHRH and nocturnal GH secretion levels. Exposure to recent stressors was not associated with GH secretion. CONCLUSIONS: Taken together with previous evidence of decreased GH after GHRH infusion in acutely depressed and recovered children, these results indicate that the decreased GH response found in high-risk subjects may represent a trait marker for depression in children and adolescents.

Analysis and classification of proton NMR spectra of lipoprotein fractions from healthy volunteers and patients with cancer or CHD.

Human blood plasma samples from 52 subjects were collected and the very low density lipoprotein (VLDL), intermediate density lipoprotein (IDL), low density lipoprotein (LDL) and high density lipoprotein were isolated by serial ultra centrifugation. 600 MHz 1H NMR spectra of the lipoprotein fractions were acquired. The methyl and methylene regions in the spectra of VLDL, LDL and HDL were utilised in further analyses via Kohonen neural networks (KNN) and generative topographic mapping (GTM), two related examples of (unsupervised learning) self-organising feature mapping techniques. Systematic variations in lipoprotein profiles can be substantially visualised through the use of KNN and GTM. The relationship between the sample positions in the Kohonen plot was visualised by surface plots of the corresponding VLDL and HDL cholesterol and VLDL triglyceride contents. The GTM maps of the VLDL and HDL fractions were used to investigate the individual properties of selected samples. A large number of the cancer patients were found clustered in the VLDL GTM map, and GTM map positions of samples in relation to CHD, diabetes and renal failure could be found. Although the study group here considered is heterogeneous in respect to age, sex, type of disease and medications within each defined class, classification of VLDL and HDL data with probabilistic neural network (PNN) was quite successful with respect to the groupings: cancer, CHD, volunteers and other (comprising patients with other diseases). Statistics based on 15 independent sets of PNN calculations gave true positive fractions usually higher than 0.83 and false positive fractions lower than 0.088. Attempts to use the corresponding LDL data and four classes were uniformly poor although some classifications (e.g., volunteer versus CHD) were easily performed.

Quantification of plasma lipids and apolipoproteins by use of proton NMR spectroscopy, multivariate and neural network analysis.

New approaches for quantification of human blood plasma lipids and apolipoproteins are presented. One method is based on multivariate analysis of proton nuclear magnetic resonance spectra of human blood plasma. Although similar approaches have been developed previously, this is the first time principal component analysis (PCA) and partial least squares regression (PLS) have been applied to this particular task. Further, a large proportion of the subjects in this study were cancer patients undergoing treatment, which introduced a new dimension to the quantification of lipoprotein distributions. Calibration models for prediction of lipids and apolipoproteins were constructed by use of PLS, and blind samples were used to test the predictive ability. Comparison of the predicted vs observed data obtained by standard clinical chemical procedures gave good agreement; the correlation coefficient for total plasma triglyceride was 0.99, for total plasma cholesterol 0.98, for LDL cholesterol 0. 97, and for HDL cholesterol 0.88. These results are comparable with those obtained with other methods. The quantitative analysis of 14 components (including total cholesterol and total triglyceride) of human blood plasma was also undertaken using various neural network (NN) analyses of selected portions of the spectra. Conventional fully connected backpropagation neural network topologies were capable of providing excellent predictions for the majority of the variables, confirming and reinforcing literature related to this approach. However HDL triglycerides were poorly predicted, while intermediate-quality results were obtained for the LDL cholesterol, plasma apoA1 and LDL apoB variables. In these instances, applying significantly different neural network algorithms involving either general regression or polynomial neural networks in combination with genetic adaptive components for parameter optimisation made improved predictions.

A pilot study of amygdala volumes in pediatric generalized anxiety disorder.

BACKGROUND: The neurodevelopment of childhood anxiety disorders is not well understood. Basic research has implicated the amygdala and circuits related to these nuclei as being central to several aspects of fear and fear-related behaviors in animals. METHODS: Magnetic resonance imaging was used to measure amygdala volumes and comparison brain regions in 12 child and adolescent subjects with generalized anxiety disorder and 24 comparison subjects. Groups were matched on age, sex, height, and handedness and were also similar on measures of weight, socioeconomic status, and full scale IQ. RESULTS: Right and total amygdala volumes were significantly larger in generalized anxiety disorder subjects. Intracranial, cerebral, cerebral gray and white matter, temporal lobe, hippocampal, and basal ganglia volumes and measures of the midsagittal area of the corpus callosum did not differ between groups. CONCLUSIONS: Although these data are preliminary and from a small sample, the results are consistent with a line of thinking that alterations in the structure and function of the amygdala may be associated with pediatric generalized anxiety disorder.

Bulk analysis of tobacco and cigarettes by magnetic resonance imaging.

Proton magnetic resonance imaging of tobacco blends in cigarette rods was investigated to assess the feasibility of various imaging protocols to characterize and quantify the structure and composition of multiphase plant materials in situ. The protocols used to characterize the rigid molecular components (plant cell wall) included single-point imaging (SPI) and a variant experiment, single-point ramped imaging with T(1) enhancement (SPRITE). Both 1D profiles, radially averaged along the length of a cigarette, and 2D maps of proton spin density and relaxation (T(2)) were acquired. Mobile components (tobacco waxes and water) were examined via conventional spin-echo imaging techniques, with 1D, 2D, and 3D data being acquired. Spin-spin relaxation times (T(2)), apparent spin-spin relaxation (T(2)), and spin-lattice (T(1)) relaxation times were measured for selected samples.

Psychosis in a pediatric mood and anxiety disorders clinic: phenomenology and correlates.

OBJECTIVES: To examine the demographics and phenomenology of psychosis in a sample of children and adolescents referred to a mood and anxiety disorders clinic. METHOD: Patients (N = 2,031) were assessed with the Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present Episode version and classified as definite, probable, or nonpsychotic. Clinical and demographic characteristics of the groups were compared,and symptoms of psychosis were analyzed using factor analysis. RESULTS: Definite psychotic symptoms were seen in approximately 90 (4.5%) patients: 80% of these reported hallucinations (mainly auditory), 22% delusions, and 3.3% thought disorder. Of the patients with definite psychotic symptoms, 24% had bipolar disorder, 41% had major depression, 21% had subsyndromal depression, and 14% had schizophrenia spectrum disorders (schizophrenia and schizoaffective disorders). Factor analysis of the definite psychotic symptoms yielded 4 factors: hallucinations, thought disorder, delusions, and manic thought disorder. Psychotic patients had a higher frequency of comorbid disorders and suicidal ideation than nonpsychotic patients. CONCLUSIONS: Outpatient youngsters with mood disorders frequently present with psychotic symptoms, in particular auditory hallucinations. These patients commonly have comorbid psychiatric disorders and suicidal ideation.

Metabolite composition in breast tumors examined by proton nuclear magnetic resonance spectroscopy.

BACKGROUND: In vivo characterisation of breast tumors using protein (1H) MR spectroscopy relies upon in vitro interpretation of tissue samples. The present study has investigated metabolite composition in extracts from breast tumors and non-involved breast tissue. Multivariate data analysis was used to determinate combinations of metabolites important for differentiation. MATERIALS AND METHODS: Tumor and non-involved breast tissue were obtained from 16 patients undergoing surgical treatment. 1H NMR spectra of perchloric acid tissue extracts were obtained at a BRUKER Avance DRX600 spectrometer. The data was analysed using principal component analysis and probabilistic neural networks. RESULTS: Low levels of glucose and high content of choline compounds were dominant findings in the tumor spectra. Principal component loadings demonstrated this strong association. The spectra were correctly classified using neural network analysis. CONCLUSIONS: Large differences in the metabolite composition of breast tumors and surrounding breast tissues have been documented.

Applications of neural network analyses to in vivo 1H magnetic resonance spectroscopy of epilepsy patients.

A total of 67 in vivo water-suppressed proton magnetic resonance spectra of the temporal lobes were recorded from 15 patients with long standing temporal lobe epilepsy and 13 healthy volunteers. Conventional data analysis indicated slightly lower N-acetyl aspartate levels in epileptic patients compared with controls. For further analysis of data, a spectral region (4.0-1.5 ppm) was used as input for artificial neural network analysis. Correct classification of spectra was obtained in 66 out of 67 cases, disregarding from which side of the brain the spectra were recorded. The ability of the trained network to recognize spectra recorded both contalaterally and ipsilaterally to the epileptic focus strongly indicates bilateral metabolic changes. Artificial neural networks could also be trained to recognize whether the spectra were recorded from the ipsilateral or contralateral side of the epileptic focus, indicating that neural network analysis of in vivo proton MR spectra can be used as an additional tool for pre-surgical lateralization of seizure foci.