Improving Alzheimer`s Disease Diagnosis with Machine Learning Techniques

There is not a specific test to diagnose Alzheimer`s disease (AD). Its diagnosis should be based upon clinical history, neuropsychological and laboratory tests, neuroimaging and electroencephalography (EEG). Therefore, new approaches are necessary to enable earlier and more accurate diagnosis and to follow treatment results. In this study we used a Machine Learning (ML) technique, named Support Vector Machine (SVM), to search patterns in EEG epochs to differentiate AD patients from controls. As a result, we developed a quantitative EEG (qEEG) processing method for automatic differentiation of patients with AD from normal individuals, as a complement to the diagnosis of probable dementia. We studied EEGs from 19 normal subjects (14 females/5 males, mean age 71.6 years) and 16 probable mild to moderate symptoms AD patients (14 females/2 males, mean age 73.4 years. The results obtained from analysis of EEG epochs were accuracy 79.9% and sensitivity 83.2%. The analysis considering the diagnosis of each individual patient reached 87.0% accuracy and 91.7% sensitivity.

THE POLICY AGENDA OF ISPRM

This paper suggests a comprehensive policy agenda and first steps to be undertaken by the International Society of Physical and Rehabilitation Medicine (ISPRM) in order to realize its humanitarian, professional and scientific mandates. The general aims of ISPRM, as formulated in its guiding documents, the relations with the World Health Organization (WHO) and the United Nations system, and demands of ISPRM`s constituency herein form the basis of this policy agenda Agenda items encompass contributions to the establishment of rehabilitation services worldwide and the development of PRM societies ISPRM`s possible input in general curricula in disability and rehabilitation, and in fighting discrimination against people experiencing disability are discuss. Moreover, the implementation of the International Classification of Functioning. Disability and Health (ICF) in medicine, contributions to WHO guidelines relevant to disability and rehabilitation the provision of a conceptual description of the rehabilitation strategy and the outline of a rehabilitation services matrix are seen as important agenda items of ISPRM`s external policy. With regard to its constituency and internal policy, a definition of the field of competence and a conceptual description of PRM, as well as the development of a consistent and comprehensive congress topic list and congress structure appear to be crucial items. The proposed agenda items serve as a basis for future discussions.

The Use of Smell Identification Tests in the Diagnosis of Parkinson`s Disease in Brazil

Smell identification tests may be of routine clinical value in the differential diagnosis of PD but are subject to cultural variation and have not been systematically evaluated in the Brazilian population. We have applied culturally adapted translations of the University of Pennsylvania 40-item smell identification test (UPSIT-40) and the 16-item identification test from Sniffin` Sticks (SS-16) to nondemented Brazilian PD patients and controls. Pearson`s correlation coefficient between the test scores was 0.76 (95% CI 0.70-0.81, n = 204, P < 0.001). To calculate reliability measures for each test we used the diagnosis (either PD or control) as outcome variable for separate logistic regression analyses using the score in the UPSIT-40 or the SS-16 as a covariate. The SS-16 specificity was 89.0% with a sensitivity of 81.1% (106 PD and 118 controls). The UPSIT-40 specificity was 83.5% and its sensitivity 82.1% (95 PD and 109 controls). Regression curves were used to associate an individual`s smell test score with the probability of belonging to the PD, as opposed to the control group. Our data provide support for the use of the UPSIT-40 and SS-16 to help distinguish early PD from controls. (c) 2008 Movement Disorder Society

Magnetic resonance imaging of idiopathic carpal tunnel syndrome: Correlation with clinical findings and electrophysiological investigation

Objective: To compare clinical evaluation, electrophysiological investigation and magnetic resonance findings in assessing the severity of idiopathic carpal tunnel syndrome. Patients and methods: Seventy-four patients with idiopathic carpal tunnel syndrome were prospectively recruited. Clinical evaluation included symptoms severity score and two-point discrimination, sensory and motor nerve conduction velocities were determined by electroneuromyography and imaging parameters were obtained after wrist magnetic resonance. The Wilcoxon test was used to define the differences between measurements of median nerve area. The Pearson and Spearman correlation tests were used to determine the relationships between all the measured parameters. Results: Cross-sectional area of median nerve was smaller at hamate level than at radio-ulnar joint and pisiform levels (p < 0.001). With exception of median nerve area at hamate level, there was a lower degree of correlation between MRI parameters and findings obtained by clinical assessments and electrophysiological measurements. The median nerve area at hamate level correlated negatively with duration of symptoms, two-point discrimination, symptoms severity score and positively with sensory nerve conduction velocity (P < 0.01). Conclusion: In patients with idiopathic carpal tunnel syndrome, median nerve area measured by wrist magnetic resonance at hamate level may be considered as a valuable indicator to grading the severity of disease. (c) 2007 Elsevier B.V. All rights reserved.

Evaluating SVM and MLDA in the extraction of discriminant regions for mental state prediction

Pattern recognition methods have been successfully applied in several functional neuroimaging studies. These methods can be used to infer cognitive states, so-called brain decoding. Using such approaches, it is possible to predict the mental state of a subject or a stimulus class by analyzing the spatial distribution of neural responses. In addition it is possible to identify the regions of the brain containing the information that underlies the classification. The Support Vector Machine (SVM) is one of the most popular methods used to carry out this type of analysis. The aim of the current study is the evaluation of SVM and Maximum uncertainty Linear Discrimination Analysis (MLDA) in extracting the voxels containing discriminative information for the prediction of mental states. The comparison has been carried out using fMRI data from 41 healthy control subjects who participated in two experiments, one involving visual-auditory stimulation and the other based on bimanual fingertapping sequences. The results suggest that MLDA uses significantly more voxels containing discriminative information (related to different experimental conditions) to classify the data. On the other hand, SVM is more parsimonious and uses less voxels to achieve similar classification accuracies. In conclusion, MLDA is mostly focused on extracting all discriminative information available, while SVM extracts the information which is sufficient for classification. (C) 2009 Elsevier Inc. All rights reserved.

Hyperplane navigation: A method to set individual scores in fMRI group datasets

Recent studies have demonstrated that spatial patterns of fMRI BOLD activity distribution over the brain may be used to classify different groups or mental states. These studies are based on the application of advanced pattern recognition approaches and multivariate statistical classifiers. Most published articles in this field are focused on improving the accuracy rates and many approaches have been proposed to accomplish this task. Nevertheless, a point inherent to most machine learning methods (and still relatively unexplored in neuroimaging) is how the discriminative information can be used to characterize groups and their differences. In this work, we introduce the Maximum Uncertainty Linear Discrimination Analysis (MLDA) and show how it can be applied to infer groups` patterns by discriminant hyperplane navigation. In addition, we show that it naturally defines a behavioral score, i.e., an index quantifying the distance between the states of a subject from predefined groups. We validate and illustrate this approach using a motor block design fMRI experiment data with 35 subjects. (C) 2008 Elsevier Inc. All rights reserved.

The impact of functional connectivity changes on support vector machines mapping of fMRI data

Functional magnetic resonance imaging (fMRI) is currently one of the most widely used methods for studying human brain function in vivo. Although many different approaches to fMRI analysis are available, the most widely used methods employ so called ""mass-univariate"" modeling of responses in a voxel-by-voxel fashion to construct activation maps. However, it is well known that many brain processes involve networks of interacting regions and for this reason multivariate analyses might seem to be attractive alternatives to univariate approaches. The current paper focuses on one multivariate application of statistical learning theory: the statistical discrimination maps (SDM) based on support vector machine, and seeks to establish some possible interpretations when the results differ from univariate `approaches. In fact, when there are changes not only on the activation level of two conditions but also on functional connectivity, SDM seems more informative. We addressed this question using both simulations and applications to real data. We have shown that the combined use of univariate approaches and SDM yields significant new insights into brain activations not available using univariate methods alone. In the application to a visual working memory fMRI data, we demonstrated that the interaction among brain regions play a role in SDM`s power to detect discriminative voxels. (C) 2008 Elsevier B.V. All rights reserved.

The gaps-in-noise test: Gap detection thresholds in normal-hearing young adults

The aim of this study was to establish parameters for the gaps-in-noise test in normal-hearing young adults. One hundred subjects (50 males and 50 females) received an audiological evaluation to rule out hearing loss and auditory processing disorder. The gaps-in-noise test was then conducted on all subjects. The mean gap detection threshold was 4.19 ms. A psychometric function by gap duration was constructed, revealing that the percentage of correct responses was less than or equal to 5% for a gap duration of 2 ms, 10-30% for a gap duration of 3 ms, 60-70% for a gap duration of 4 ms, and over 96% for gap durations of 5 ms or longer. The results suggest that the data obtained can be applied as reference values for future testing. In the subjects evaluated, the gaps-in-noise test proved to be consistent with low variability.

Auditory results from cochlear implants in elderly people

According to data from the Brazilian Institute of Geography and Statistics, the elderly population grew 47.8% in the last decade in Brazil. A portion of this population has severe and/or profound hearing loss and do not benefit from conventional hearing aids. Thus, the use of cochlear implant is required. Aim: To analyze the benefits of cochlear implants in the elderly based on the comparison of primary auditory thresholds before and after the operation, discrimination of sentences in speech and in talking on the telephone. Methodology: Retrospective cohort study, analyzing medical records from patients aged over 60 years, users of cochlear implant for at least 1 year. Results: Fourteen medical records were analyzed. Mean age of patients was 63.07 years. The mean pure tone thresholds between 500Hz, 1kHz, 2kHz and 4kHz before the implantation was 113dBHL. None of the patients, before operation, could discriminate sentences in open sets and only 3 scored 17% in closed sets sentence recognition. After one year of implantation, the mean sound field thresholds reached 34dBHL, and open set sentences recognition of 93.57%, while 71% of the patients had become able to have a conversation on the telephone. Conclusion: The elderly users of cochlear implant showed important outcomes, with significant improvement in understanding in the open set and in using the telephone.

Comparison of Fourier-Domain and Time-Domain Optical Coherence Tomography in the Detection of Band Atrophy of the Optic Nerve

PURPOSE: To compare the ability of Fourier-domain (FD) optical coherence tomography (3D OCT-1000; Top, con, Tokyo, Japan) and time domain (TD) OCT (Stratus; Carl Zeiss Meditec Inc, Dublin, California, USA) to detect axonal loss in eyes with band atrophy (BA) of the optic nerve. DESIGN: Cross-sectional study. METHODS: Thirty-six eyes from 36 patients with BA and temporal visual field (VF) defect from chiasmal compression and 36 normal eyes were studied. Subjects were submitted to standard automated perimetry and macular and retinal nerve fiber layer (RNFL) measurements were taken using 3D OCT-1000 and Stratus OCT. Receiver operating characteristic (ROC) curves were calculated for each parameter. Spearman correlation coefficients were obtained to evaluate the relationship between RNFL and macular thickness parameters and severity of VF loss. Measurements from the two devices were compared. RESULTS: Regardless of OCT device, all RNFL and macular thickness parameters were significantly lower in eyes with BA compared with normal eyes, but no statistically significant difference was found with regard to the area under the ROC curve. Structure-function relationships were also similar for the two devices. In both groups, RNFL and macular thickness measurements were generally and in some cases significantly smaller with 3D OCT-1000 than with Stratus OCT. CONCLUSIONS: The introduction of FD technology did not lead to better discrimination ability for detecting BA of the optic nerve compared with TD technology when using the software currently provided by the manufacturer. 3D OCT-1000 FD OCT RNFL and macular measurements were generally smaller than TD Stratus OCT measurements. Investigators should be aware of this fact when comparing measurements obtained with these two devices. (Am J Oplathalmol 2009;147: 56-63. (c) 2009 by Elsevier Inc. All rights reserved.)

Increased Melanocytic Nevi in Patients with Inherited Ichthyoses: Report of a Previously Undescribed Association

Ichthyosis is a heterogeneous cornification disorder. Melanocytic lesions have not been previously described in association with ichthyosis. Their clinical importance lies in the fact that they may simulate melanoma clinically and dermoscopically, as seen in epidermolysis bullosa. The objective of this study was to evaluate the clinical, dermoscopic, and histopathologic features of nevi and lentigines in 16 patients with autosomal recessive congenital ichthyosis-lamellar ichthyosis and non-bullous ichthyosiform congenital erythroderma. Patients underwent general clinical examination dermoscopy. The more suspicious lesions were excised and to histopathologic examination. Most patients (n = 13) reported no personal or familial history of melanoma or atypical nevi. All of the patients had at least five atypical melanocytic lesions. Ten of the 16 patients had at least one atypical nevus or lentigo. This study suggests that increased atypical melanocytic nevi may be a feature of long-standing congenital ichthyoses. Whether this finding is disease-related or a coincidental observation is difficult to ascertain. As an unequivocal discrimination from malignant melanoma in vivo is not always possible, regular clinical follow-up of patients with ichthyosis and increased or unusual nevi is recommended.

Effect of brain-derived neurotrophic factor Val66Met polymorphism and serum levels on the progression of mild cognitive impairment

Objectives. Abnormalities in neurotrophic systems have been reported in Alzheimer`s disease (AD), as shown by decreased serum brain-derived neurotrophic factor (BDNF) levels and association with BDNF genetic polymorphisms. In this study, we investigate whether these findings can be detected in patients with mild cognitive impairment (MCI), which is recognized as a high risk condition for AD. We also address the impact of these variables on the progression of cognitive deficits within the MCI-AD continuum. Methods. One hundred and sixty older adults with varying degrees of cognitive impairment (30 patients with AD, 71 with MCI, and 59 healthy controls) were longitudinally assessed for up to 60 months. Baseline serum BDNF levels were determined by sandwich ELISA, and the presence of polymorphisms of BDNF and apolipoprotein E (Val66Met and APOE*E4, respectively) was determined by allelic discrimination analysis on real time PCR. Modifications of cognitive state were ascertained for non-demented subjects. Results. Mean serum BDNF levels were reduced in patients with MCI and AD, as compared to controls (509.2 +/- 210.5; 581.9 +/- 379.4; and 777.5 +/- 467.8 pg/l respectively; P < 0.001). Baseline serum BDNF levels were not associated with the progression of cognitive impairment upon follow-up in patients with MCI (progressive MCI, 750.8 +/- 463.0; stable MCI, 724.0 +/- 343.4; P = 0.8), nor with the conversion to AD. Although Val66Met polymorphisms were not associated with the cross-sectional diagnoses of MCI or AD, the presence of Met-BDNF allele was associated with a higher risk of disease-progression in patients with MCI (OR = 3.0 CI(95%) [1.2-7.8], P = 0.02). We also found a significant interaction between the APOE*E4 and Met-BDNF allele increasing the risk of progression of cognitive impairment in MCI patients (OR = 4.4 CI(95%) [1.6-12.1], P = 0.004). Conclusion. Decreased neurotrophic support, as indicated by a reduced systemic availability of BDNF, may play role in the neurodegenerative processes that underlie the continuum from MCI to AD. The presence of Met-BDNF allele, particularly in association with APOE*E4, may predict a worse cognitive outcome in patients with MCI.

Associations of serious mental illness with earnings: results from the WHO World Mental Health surveys

Background Burden-of-illness data, which are often used in setting healthcare policy-spending priorities, are unavailable for mental disorders in most countries. Aims To examine one central aspect of illness burden, the association of serious mental illness with earnings, in the World Health Organization (WHO) World Mental Health (WMH) Surveys. Method The WMH Surveys were carried out in 10 high-income and 9 low- and middle-income countries. The associations of personal earnings with serious mental illness were estimated. Results Respondents with serious mental illness earned on average a third less than median earnings, with no significant between-country differences (chi(2)(9)=5.5-8.1, P=0.5-0.79). These losses are equivalent to 0.3-0.8% of total national earnings. Reduced earnings among those with earnings and the increased probability of not earning are both important components of these associations: Conclusions These results add to a growing body of evidence that mental disorders have high societal costs. Decisions about healthcare resource allocation should take these costs into consideration.

Myocardial Perfusion Imaging Is a Strong Predictor of Death in Women

OBJECTIVES We sought to assess the prognostic value and risk classification improvement using contemporary single-photon emission computed tomography myocardial perfusion imaging (SPECT-MPI) to predict all-cause mortality. BACKGROUND Myocardial perfusion is a strong estimator of prognosis. Evidence published to date has not established the added prognostic value of SPECT-MPI nor defined an approach to detect improve classification of risk in women from a developing nation. METHODS A total of 2,225 women referred for SPECT-MPI were followed by a mean period of 3.7 +/- 1.4 years. SPECT-MPI results were classified as abnormal on the presence of any perfusion defect. Abnormal scans were further classified as with mild/moderate reversible, severe reversible, partial reversible, or fixed perfusion defects. Risk estimates for incident mortality were categorized as <1%/year, 1% to 2%/year, and >2%/year using Cox proportional hazard models. Risk-adjusted models incorporated clinical risk factors, left ventricular ejection fraction (LVEF), and perfusion variables. RESULTS All-cause death occurred in 139 patients. SPECT-MPI significantly risk stratified the population; patients with abnormal scans had significantly higher death rates compared with patients with normal scans, 13.1% versus 4.0%, respectively (p < 0.001). Cox analysis demonstrated that after adjusting for clinical risk factors and LVEF, SPECT-MPI improved the model discrimination (integrated discrimination index = 0.009; p = 0.02), added significant incremental prognostic information (global chi-square increased from 87.7 to 127.1; p < 0.0001), and improved risk prediction (net reclassification improvement = 0.12; p = 0.005). CONCLUSIONS SPECT-MPI added significant incremental prognostic information to clinical and left ventricular functional variables while enhancing the ability to classify this Brazilian female population into low-and high-risk categories of all-cause mortality. (J Am Coll Cardiol Img 2011;4:880-8) (C) 2011 by the American College of Cardiology Foundation

Verbal fluency in the detection of mild cognitive impairment and Alzheimer`s disease among Brazilian Portuguese speakers: the influence of education

Background: Verbal fluency (VF) tasks are simple and efficient clinical tools to detect executive dysfunction and lexico-semantic impairment. VF tasks are widely used in patients with suspected dementia, but their accuracy for detection of mild cognitive impairment (MCI) is still under investigation. Schooling in particular may influence the subject`s performance. The aim of this study was to compare the accuracy of two semantic categories (animals and fruits) in discriminating controls, MCI patients and Alzheimer`s disease (AD) patients. Methods: 178 subjects, comprising 70 controls (CG), 70 MCI patients and 38 AD patients, were tested on two semantic VF tasks. The sample was divided into two schooling groups: those with 4-8 years of education and those with 9 or more years. Results: Both VF tasks - animal fluency (VFa) and fruits fluency (VFf) - adequately discriminated CG from AD in the total sample (AUC = 0.88 +/- 0.03, p < 0.0001) and in both education groups, and high educated MCI from AD (VFa: AUC = 0.82 +/- 0.05, p < 0.0001; VFf: AUC = 0.85 +/- 0.05, p < 0.0001). Both tasks were moderately accurate in discriminating CG from MCI (VFa: AUC = 0.68 +/- 0.04, p < 0.0001 - VFf:AUC = 0.73 +/- 0.04, p < 0.0001) regardless of the schooling level, and MCI from AD in the total sample (VFa: AUC = 0.74 +/- 0.05, p < 0.0001; VFf: AUC = 0.76 +/- 0.05, p < 0.0001). Neither of the two tasks differentiated low educated MCI from AD. In the total sample, fruits fluency best discriminated CG from MCI and MCI from AD; a combination of the two improved the discrimination between CG and AD. Conclusions: Both categories were similar in discriminating CG from AD; the combination of both categories improved the accuracy for this distinction. Both tasks were less accurate in discriminating CG from MCI, and MCI from AD.