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.

Normal and anomalous classification of category exemplars in schizophrenia

Background Schizophrenia has been associated with semantic memory impairment and previous studies report a difficulty in accessing semantic category exemplars (Moelter et al. 2005 Schizophr Res 78:209–217). The anterior temporal cortex (ATC) has been implicated in the representation of semantic knowledge (Rogers et al. 2004 Psychol Rev 111(1):205–235). We conducted a high-field (4T) fMRI study with the Category Judgment and Substitution Task (CJAST), an analogue of the Hayling test. We hypothesised that differential activation of the temporal lobe would be observed in schizophrenia patients versus controls. Methods Eight schizophrenia patients (7M : 1F) and eight matched controls performed the CJAST, involving a randomised series of 55 common nouns (from five semantic categories) across three conditions: semantic categorisation, anomalous categorisation and word reading. High-resolution 3D T1-weighted images and GE EPI with BOLD contrast and sparse temporal sampling were acquired on a 4T Bruker MedSpec system. Image processing and analyses were performed with SPM2. Results Differential activation in the left ATC was found for anomalous categorisation relative to category judgment, in patients versus controls. Conclusions We examined semantic memory deficits in schizophrenia using a novel fMRI task. Since the ATC corresponds to an area involved in accessing abstract semantic representations (Moelter et al. 2005), these results suggest schizophrenia patients utilise the same neural network as healthy controls, however it is compromised in the patients and the different ATC activity might be attributable to weakening of category-to-category associations.

A population-based morphometric MRI study in patients with first-episode psychotic bipolar disorder: comparison with geographically matched healthy controls and major depressive disorder subjects

Objectives: Many morphometric magnetic resonance imaging (MRI) studies that have investigated the presence of gray matter (GM) volume abnormalities associated with the diagnosis of bipolar disorder (BD) have reported conflicting findings. None of these studies has compared patients with recent-onset psychotic BD with asymptomatic controls selected from exactly the same environment using epidemiological methods, or has directly contrasted BD patients against subjects with first-onset psychotic major depressive disorder (MDD). We examined structural brain differences between (i) BD (type I) subjects and MDD subjects with psychotic features in their first contact with the healthcare system in Brazil, and (ii) these two mood disorder groups relative to a sample of geographically matched asymptomatic controls. Methods: A total of 26 BD subjects, 20 subjects with MDD, and 94 healthy controls were examined using either of two identical MRI scanners and acquisition protocols. Diagnoses were based on DSM-IV criteria and confirmed one year after brain scanning. Image processing was conducted using voxel-based morphometry. Results: The BD group showed increased volume of the right dorsal anterior cingulate cortex relative to controls, while the MDD subjects exhibited bilateral foci GM deficits in the dorsolateral prefrontal cortex (p < 0.05, corrected for multiple comparisons). Direct comparison between BD and MDD patients showed a focus of GM reduction in the right-sided dorsolateral prefrontal cortex (p < 0.05, corrected for multiple comparisons) and a trend (p < 0.10, corrected) toward left-sided GM deficits in the dorsolateral prefrontal cortex of MDD patients. When analyses were repeated with scanner site as a confounding covariate the finding of increased right anterior cingulate volumes in BD patients relative to controls remained statistically significant (p = 0.01, corrected for multiple comparisons). Conclusions: These findings reinforce the view that there are important pathophysiological distinctions between BD and MDD, and indicate that subtle dorsal anterior cingulate abnormalities may be relevant to the pathophysiology of BD.

Palpebral fissure changes after monobloc frontofacial advancement in faciocraniosynostosis

There is virtually no literature on the effect of correction of syndromal faciocraniosynostosis with monobloc advancement on the palpebral fissure shape. Using image processing software, we measured the effect of monobloc advancement on the position of the upper and lower eyelids as well as the palpebral fissure slant in a series of 18 patients with syndromal faciocraniosynostosis who had undergone surgery for correction of orbital and midface hypoplasia. For both eyed of each patient, 3 variables were measured on the pre- and postoperative photographs: the linear distances between the upper and lower eyelid margins, the pupil center and the angle between the inner and outer canthi. The globe protrusion was also measured on axial computed tomography scans before and after surgery. The results indicate that the exorbitism reduction induced by monobloc advancement is accompanied by a diminution of the distance between both eyelids and the pupil center. However, the downward slant of the palpebral fissure is increased after surgery. The data suggest that the lower and upper eyelid retraction seen preoperatively in the majority of patients with faciocraniosynostosis tend to be corrected when the frontofacial region is advanced by the monobloc. On the other hand, the surgery tends to lower the outer canthus, increasing the negative slant of the fissure. The postoperative changes induced by the frontofacial monobloc advancement need to be taken into account when the surgery is going to be performed.

Subcritical crack growth in porcelains, glass-ceramics, and glass-infiltrated alumina composite for dental restorations

The objective was to compare fracture toughness (K(Ic)), stress corrosion susceptibility coefficient (n), and stress intensity factor threshold for crack propagation (K(I0)) of two porcelains [VM7/Vita (V) and d.Sign/Ivoclar (D)], two glass-ceramics [Empress/Ivolcar (E1) and Empress2/Ivlocar (E2)] and a glass-infiltrated alumina composite [In-Ceram Alumina/Vita (IC)]. Disks were constructed according to each manufacturer`s processing method, and polished before induction of cracks by a Vickers indenter. Crack lengths were measured under optical microscopy at times between 0.1 and 100 h. Specimens were stored in artificial saliva at 37A degrees C during the whole experiment. K(Ic) and n were determined using indentation fracture method. K(I0) was determined by plotting log crack velocity versus log K(I). Microstructure characterization was carried out under SEM, EDS, X-ray diffraction and X-ray fluorescence. IC and E2 presented higher K(Ic) and K(I0) compared to E1, V, and D. IC presented the highest n value, followed by E2, D, E1, and V in a decreasing order. V and D presented similar K(Ic), but porcelain V showed higher K(I0) and lower n compared to D. Microstructure features (volume fraction, size, aspect ratio of crystalline phases and chemical composition of glassy matrix) determined K(Ic). The increase of K(Ic) value favored the increases of n and K(I0).

Using a cartographic modeling language to manipulate spectral satellite imagery

Land related information about the Earth's surface is commonIJ found in two forms: (1) map infornlation and (2) satellite image da ta. Satellite imagery provides a good visual picture of what is on the ground but complex image processing is required to interpret features in an image scene. Increasingly, methods are being sought to integrate the knowledge embodied in mop information into the interpretation task, or, alternatively, to bypass interpretation and perform biophysical modeling directly on derived data sources. A cartographic modeling language, as a generic map analysis package, is suggested as a means to integrate geographical knowledge and imagery in a process-oriented view of the Earth. Specialized cartographic models may be developed by users, which incorporate mapping information in performing land classification. In addition, a cartographic modeling language may be enhanced with operators suited to processing remotely sensed imagery. We demonstrate the usefulness of a cartographic modeling language for pre-processing satellite imagery, and define two nerv cartographic operators that evaluate image neighborhoods as post-processing operations to interpret thematic map values. The language and operators are demonstrated with an example image classification task.

Theory and applications of fuzzy Volterra integral equations

Formulations of fuzzy integral equations in terms of the Aumann integral do not reflect the behavior of corresponding crisp models. Consequently, they are ill-adapted to describe physical phenomena, even when vagueness and uncertainty are present. A similar situation for fuzzy ODEs has been obviated by interpretation in terms of families of differential inclusions. The paper extends this formalism to fuzzy integral equations and shows that the resulting solution sets and attainability sets are fuzzy and far better descriptions of uncertain models involving integral equations. The investigation is restricted to Volterra type equations with mildly restrictive conditions, but the methods are capable of extensive generalization to other types and more general assumptions. The results are illustrated by integral equations relating to control models with fuzzy uncertainties.

High resolution 3-D imaging via multi-pass SAR

Spaceborne/airborne synthetic aperture radar (SAR) systems provide high resolution two-dimensional terrain imagery. The paper proposes a technique for combining multiple SAR images, acquired on flight paths slightly separated in the elevation direction, to generate high resolution three-dimensional imagery. The technique could be viewed as an extension to interferometric SAR (InSAR) in that it generates topographic imagery with an additional dimension of resolution. The 3-D multi-pass SAR imaging system is typically characterised by a relatively short ambiguity length in the elevation direction. To minimise the associated ambiguities we exploit the relative phase information within the set of images to track the terrain landscape. The SAR images are then coherently combined, via a nonuniform DFT, over a narrow (in elevation) volume centred on the 'dominant' terrain ground plane. The paper includes a detailed description of the technique, background theory, including achievable resolution, and the results of an experimental study.

Mapping glider habitat in dry eucalypt forests for Montreal Process indicator 1.1: Fragmentation of forest types

Accurate habitat mapping is critical to landscape ecological studies such as required for developing and testing Montreal Process indicator 1.1e, fragmentation of forest types. This task poses a major challenge to remote sensing, especially in mixedspecies, variable-age forests such as dry eucalypt forests of subtropical eastern Australia. In this paper, we apply an innovative approach that uses a small section of one-metre resolution airborne data to calibrate a moderate spatial resolution model (30 m resolution; scale 1:50 000) based on Landsat Thematic Mapper data to estimate canopy structural properties in St Marys State Forest, near Maryborough, south-eastern Queensland. The approach applies an image-processing model that assumes each image pixel is significantly larger than individual tree crowns and gaps to estimate crown-cover percentage, stem density and mean crown diameter. These parameters were classified into three discrete habitat classes to match the ecology of four exudivorous arboreal species (yellowbellied glider Petaurus australis, sugar glider P. breviceps, squirrel glider P. norfolcensis , and feathertail glider Acrobates pygmaeus), and one folivorous arboreal marsupial, the greater glider Petauroides volans. These species were targeted due to the known ecological preference for old trees with hollows, and differences in their home range requirements. The overall mapping accuracy, visually assessed against transects (n = 93) interpreted from a digital orthophoto and validated in the field, was 79% (KHAT statistic = 0.72). The KHAT statistic serves as an indicator of the extent that the percentage correct values of the error matrix are due to ‘true’ agreement verses ‘chance’ agreement. This means that we are able to reliably report on the effect of habitat loss on target species, especially those with a large home range size (e.g. yellow-bellied glider). However, the classified habitat map failed to accurately capture the spatial patterning (e.g. patch size and shape) of stands with a trace or sub-dominance of senescent trees. This outcome makes the reporting of the effects of habitat fragmentation more problematic, especially for species with a small home range size (e.g. feathertail glider). With further model refinement and validation, however, this moderateresolution approach offers an important, cost eff e c t i v e advancement in mapping the age of dry eucalypt forests in the region.

Monitoring the composition and form of urban environments based on the vegetation-impervious surface-soil (VIS) model by sub-pixel analysis techniques

The majority of the world's population now resides in urban environments and information on the internal composition and dynamics of these environments is essential to enable preservation of certain standards of living. Remotely sensed data, especially the global coverage of moderate spatial resolution satellites such as Landsat, Indian Resource Satellite and Systeme Pour I'Observation de la Terre (SPOT), offer a highly useful data source for mapping the composition of these cities and examining their changes over time. The utility and range of applications for remotely sensed data in urban environments could be improved with a more appropriate conceptual model relating urban environments to the sampling resolutions of imaging sensors and processing routines. Hence, the aim of this work was to take the Vegetation-Impervious surface-Soil (VIS) model of urban composition and match it with the most appropriate image processing methodology to deliver information on VIS composition for urban environments. Several approaches were evaluated for mapping the urban composition of Brisbane city (south-cast Queensland, Australia) using Landsat 5 Thematic Mapper data and 1:5000 aerial photographs. The methods evaluated were: image classification; interpretation of aerial photographs; and constrained linear mixture analysis. Over 900 reference sample points on four transects were extracted from the aerial photographs and used as a basis to check output of the classification and mixture analysis. Distinctive zonations of VIS related to urban composition were found in the per-pixel classification and aggregated air-photo interpretation; however, significant spectral confusion also resulted between classes. In contrast, the VIS fraction images produced from the mixture analysis enabled distinctive densities of commercial, industrial and residential zones within the city to be clearly defined, based on their relative amount of vegetation cover. The soil fraction image served as an index for areas being (re)developed. The logical match of a low (L)-resolution, spectral mixture analysis approach with the moderate spatial resolution image data, ensured the processing model matched the spectrally heterogeneous nature of the urban environments at the scale of Landsat Thematic Mapper data.