Gísli Hólmar Jóhannesson

Diagnostic Accuracy of Statistical Pattern Recognition of Electroencephalogram Registration in Evaluation of Cognitive Impairment and Dementia

Background: There is still a need for simple, noninvasive, and inexpensive methods to diagnose the causes of cognitive impairment and dementia. In this study, contemporary statistical methods were used to classify the clinical cases of cognitive impairment based on electroencephalograms (EEG). Methods: An EEG database was established from seven different groups of subjects with cognitive impairment and dementia as well as healthy controls. A classifier was created for each possible pair of groups using statistical pattern recognition (SPR). Results: A good-to-excellent separation was found when differentiating cases of degenerative disorders from controls, vascular disorders, and depression but this was less so when the likelihood of comorbidity was high. Conclusions: Using EEG with SPR seems to be a reliable method for diagnosing the causes of cognitive impairment and dementia, but comorbidity must betaken into account.

The Acetylcholine Index: An Electroencephalographic Marker of Cholinergic Activity in the Living Human Brain Applied to Alzheimer's Disease and Other Dementias

Background: The cholinergic hypothesis is well established and has led to the development of pharmacological treatments for Alzheimers disease (AD). However, there has previously been no physiological means of monitoring cholinergic activity in vivo. Methods: An electroencephalography (EEG)-based acetylcholine (Ach) index reflecting the cholinergic activity in the brain was developed using data from a scopolamine challenge study. The applicability of the Ach index was examined in an elderly population of healthy controls and patients suffering from various causes of cognitive decline. Results: The Ach index showed a strong reduction in the severe stages of AD dementia. A high correlation was demonstrated between the Ach index and cognitive function. The index was reduced in patients with mild cognitive impairment and prodromal AD, indicating a decreased cholinergic activity. When considering the distribution of the Ach index in a population of healthy elderly subjects, an age-related threshold was revealed, beyond which there is a general decline in cholinergic activity. Conclusions: The EEG-based Ach index provides, for the first time, a physiological means of monitoring the cholinergic activity in the human brain in vivo. This has great potential for aiding diagnosis and patient stratification as well as for monitoring disease progression and treatment response.

Multimodal EEG-MRI in the differential diagnosis of Alzheimer's disease and dementia with Lewy bodies.

Differential diagnosis of Alzheimers disease (AD) and dementia with Lewy bodies (DLB) remains challenging; currently the best discriminator is striatal dopaminergic imaging. However this modality fails to identify 15–20% of DLB cases and thus other biomarkers may be useful. It is recognised electroencephalography (EEG) slowing and relative medial temporal lobe preservation are supportive features of DLB, although individually they lack diagnostic accuracy. Therefore, we investigated whether combined EEG and MRI indices could assist in the differential diagnosis of AD and DLB. Seventy two participants (21 Controls, 30 AD, 21 DLB) underwent resting EEG and 3 T MR imaging. Six EEG classifiers previously generated using support vector machine algorithms were applied to the present dataset. MRI index was derived from medial temporal atrophy (MTA) ratings. Logistic regression analysis identified EEG predictors of AD and DLB. A combined EEG-MRI model was then generated to examine whether there was an improvement in classification compared to individual modalities. For EEG, two classifiers predicted AD and DLB (model: χ2 = 22.1, df = 2, p < 0.001, Nagelkerke R2 = 0.47, classification = 77% (AD 87%, DLB 62%)). For MRI, MTA also predicted AD and DLB (model: χ2 = 6.5, df = 1, p = 0.01, Nagelkerke R2 = 0.16, classification = 67% (77% AD, 52% DLB). However, a combined EEG-MRI model showed greater prediction in AD and DLB (model: χ2 = 31.1, df = 3, p < 0.001, Nagelkerke R2 = 0.62, classification = 90% (93% AD, 86% DLB)). While suggestive and requiring validation, diagnostic performance could be improved by combining EEG and MRI, and may represent an alternative to dopaminergic imaging.

Electroencephalography as a clinical tool for diagnosing and monitoring attention deficit hyperactivity disorder: a cross-sectional study

Objectives The aim of this study was to develop and test, for the first time, a multivariate diagnostic classifier of attention deficit hyperactivity disorder (ADHD) based on EEG coherence measures and chronological age. Setting The participants were recruited in two specialised centres and three schools in Reykjavik. Participants The data are from a large cross-sectional cohort of 310 patients with ADHD and 351 controls, covering an age range from 5.8 to 14 years. ADHD was diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders fourth edition (DSM-IV) criteria using the K-SADS-PL semistructured interview. Participants in the control group were reported to be free of any mental or developmental disorders by their parents and had a score of less than 1.5 SDs above the age-appropriate norm on the ADHD Rating Scale-IV. Other than moderate or severe intellectual disability, no additional exclusion criteria were applied in order that the cohort reflected the typical cross section of patients with ADHD. Results Diagnostic classifiers were developed using statistical pattern recognition for the entire age range and for specific age ranges and were tested using cross-validation and by application to a separate cohort of recordings not used in the development process. The age-specific classification approach was more accurate (76% accuracy in the independent test cohort; 81% cross-validation accuracy) than the age-independent version (76%; 73%). Chronological age was found to be an important classification feature. Conclusions The novel application of EEG-based classification methods presented here can offer significant benefit to the clinician by improving both the accuracy of initial diagnosis and ongoing monitoring of children and adolescents with ADHD. The most accurate possible diagnosis at a single point in time can be obtained by the age-specific classifiers, but the age-independent classifiers are also useful as they enable longitudinal monitoring of brain function.

Quantitative eeg applying the statistical pattern recognition method: A useful tool in the dementia diagnostic work-up

Background:MicroRNAs are implicated in the pathogenesis of Alzheimer’s disease (AD). However, the change of microRNA expression and the impact of dys-regulated microRNAs in AD remains unclear. Methods:Using CHIP technique we screened the microRNAs which were dys-regulated in hippocampus of transgenic mice of AD, and using a sensitive qRT-PCR platform we further validated the differentially expressed microRNAs. Results: The data revealed a total of 5 miRNAs (cut-off>1⁄42.0 folds) were differentially expressed in hippocampus of AD transgenic mice, including four up-regulated miRNAs, miR-34a, miR-5625, miR5130 and miR-1950, and one down-regulated miRNA, miR-882. Conclusions: Combined with literature and bioinformatics analysis, the data revealed how the deregulated brain microRNAs may be involved in the pathological processes and neurobehavioral abnormalities such as amyloid processing, synaptic loss and anxiety, and memory loss. Acknowledgements: This work was supported by Shenzhen Special Fund Project on Strategic Emerging Industry Development (JCYJ20130329103949650), Medical Scientific Research Foundation of Guangdong Province (A2013598), and Shenzhen Scheme of Science and Technology (Medicine and Health) (201302148).

EEG phenotyping and sub-phenotyping of Alzheimer's disease and other dementias

internationally accepted criteria, and AD cases were further classified for the presence of cerebrovascular disease (CBVD). Controls with no history of stroke and cognitive impairment were selected from the Singapore Epidemiology of Eye Disease program and matched by race, gender and 5year age groups. Retinal vascular parameters (retinal vascular caliber, fractal dimension and tortuosity) were assessed using a semi-automated computer-based program. Due to its skewed distribution, retinal vessel tortuosity was log-transformed. Logistic regression models were constructed adjusting for gender, age, hypertension, diabetes and hypercholesterolemia status. Results: A total of 156 AD cases (98 without CBVD and 58with CBVD), 27 vascular dementia cases, and 493 controls were included in this preliminary analysis. Narrower arteriolar caliber was associated with VaD (multivariable-adjusted odds ratio (OR) per standard deviation (SD) decrease: 2.41; 95%CI: 1.13-5.14) and AD with CBVD (OR per SD decrease: 1.84; 95%CI: 1.05-3.23). Narrower venular caliber (OR per SD decrease: 1.72; 95%CI: 1.11-2.68), decreased arteriolar fractal dimension (OR per SD decrease: 1.29; 95%CI: 1.03-1.61) and venular fractal dimension (OR per SD increase: 1.36; 95%CI: 1.08-1.72) were associated with only AD without CBVD. However, increased arteriolar and venular tortuosity was associated with all three subtypes of dementia. Conclusions: A sparser retinal microvascular network was associated with AD, whereas narrower arterioles were associated with dementia linked to CBVD, and tortuous retinal vessels were associated with all three subtypes of dementia. This suggests that retinal vascular parameters may potentially useful in assessing the contribution of microvascular pathology to the different subtpyes of dementia.

Treatment response prediction in people with Alzheimer's disease using EEG

Background: The majority of individuals referred to a Memory Clinic have subtle symptoms of cognitive impairment and from a clinical point of view most of them are not demented. Eventually, many of these individuals develop a form of dementia that is clinically detectable but some of them may remain stable for years, and the concern may even disappear. The EEG diagnostic tool presented here is designed to help the clinician to evaluate possible progress of the cognitive symptoms in MCI subjects over time. Methods: In an EEG database EEGs from 342 mild AD subjects and 181 MCI individuals have been entered. The EEG data from the MCI individuals was collected over a period of 7 years (20052011). Over the course of 8 years these MCI individuals have been followed up. The MCI individuals were divided into three subgroups according to their clinical status at follow up: prodromal AD (pAD), stable MCI (sMCI), and other. The pAD and sMCI groups were compared by constructing a classifier by applying statistical pattern recognition to a large set of EEG features. Results: Of the 181 MCI individuals 70 were diagnosed with AD 1-7.7 years after the EEG measurement with a mean of 2 years. 79 remainedMCI for at least 3 years, with a mean of 4.75 years and standard deviation of 1.00 year. The remaining 32 individuals had developed other types of dementia at follow up. The pAD-sMCI classifier indicates that using EEG it is possible to predict which of the MCI individuals will at a later time develop AD with accuracy of about 80%. Conclusions: Following amore substantial clinical validation and an easy access to the methodology, we expect this application of clinical EEG in support for differential diagnosis of mild cognitive impairment to become a realistic first step in the full clinical workup of patients who visit a memory clinic. The underlying technology is well known, widely available and inexpensive in relation to other imaging techniques.

IC-P2-106: Diagnosing, tracking and evaluation of treatment response for Alzheimer's disease using EEG and database- supported diagnostics

and mild cognitive impairment (MCI) in comparison with healthy subjects. The most prominent change is the reduction of the neuronal marker N-acetylaspartate (NAA). Recent studies have shown that low NAA levels predict cognitive decline in MCI. Furthermore, NAA increases during acetylcholinesterase-inhibitor treatment in AD. Thus, 1H-MRS measures may be candidates for a biomarkers or surrogate markers in clinical trials. A prerequisite for the application of 1H-MRS in large clinical trials is the multicenter feasibility. Methods: Within the German Competence Network on Dementia, we performed the first 1H-MRS multicenter study of the medial temporal lobe, as a region of particular interest in AD and MCI patients. The study was conducted at four German sites. We measured metabolic ratios and quantities of different molecular groups (NAA, choline compounds, creatine/phosophcreatine, myoinositol, glutamine/glutamate). In the presentation, the data from the crossectional study will be reported. The sample included 125 patients with dementia, 123 MCI patients and 48 comparison subjects. Results: We observed center effects for different metabolites and ratios. Z-transformed data, however, showed difference between diagnostic groups. A main result is a significant group effect on the metabolic ratio NAA/Cr with AD patients showing significantly lower measures than the healthy comparison subjects and MCI patients showing measures in between. The detailed results will be given in the presentation. Conclusions: Our study shows that multicenter 1H-MRS of the medial temporal lobe in patients with mild AD and MCI can be performed, which is the basis of 1H-MRS as a biomarker or surrogate marker tool in clinical trials.

P-040: Database supported diagnostics of Alzheimer’s disease

Background: The possibility of using electroencephalograms (EEG) as a surrogate marker for Alzheimer’s disease (AD) has been indicated by various investigations. Although the EEG of AD patients on average differs from healthy individuals for a given feature of the EEG, using a single feature is not clinically useful since the overlap between the distributions of the feature between the groups is too great resulting in poor accuracy. Objective: In this project a database of EEGs and multiple EEG features are used for diagnostic purposes for AD. The main goal is to create a diagnostic tool for AD which is both sensitive and specific. A secondary objective is to investigate whether AD patients can be separated from patients suffering from Vascular Dementia (VD) using the same methodology. Methods: A database of 1000 individual EEG measurements is being constructed. The participant has eyes closed and is at rest and the EEG is measured for a few minutes. Currently 500 EEGs have been collected. The results of EEG measurements from AD patients, VD patients, and age matched controls are presented here. Statistical pattern recognition (SPR) is applied to the dataset in order to select the combination of EEG features which best separates the groups of AD patients and healthy individuals and the groups of AD patients and VD patients. Results: The AD group and the control group can be separated with 90% accuracy. And the area under the receiver operator curve (ROC) is 0.96 demonstrating the applicability of the method. Conclusions: The results indicate that this type of methodology is both accurate and specific and can be used for a detection of AD. It may therefore be a useful addition to traditional AD diagnostic procedure, in particular since the method is completely objective.