913 resultados para INDEPENDENT COMPONENT ANALYSIS (ICA)
Resumo:
Le vieillissement étant un enjeu démographique majeur, il est capital de mieux comprendre les changements qui surviennent durant cette période de la vie. Il est connu que certaines fonctions cognitives sont modifiées avec l’avancée en âge. D’ailleurs, 25 à 50 % des personnes âgées de 65 ans et plus rapportent avoir observé un déclin de leur cognition et de leur mémoire. Les travaux de cette thèse portent sur la caractérisation de la plainte cognitive chez des personnes âgées saines et chez des aînés ayant un trouble cognitif léger (TCL) ainsi que sur son évolution au fil de la progression vers la maladie d’Alzheimer. La première étude (Chapitre II) avait pour objectif d’identifier les différents domaines de plainte mnésique chez des personnes d’âge moyen et des individus plus âgés. Elle visait également à vérifier si les domaines de plainte étaient associés aux performances aux tests neuropsychologiques. L’effet sur la plainte de certaines caractéristiques personnelles (âge, sexe, niveau de scolarité et symptômes dépressifs) a aussi été examiné. Le Questionnaire d’auto-évaluation de la mémoire (QAM; Van der Linden, Wijns, Von Frenkell, Coyette, & Seron, 1989) et plusieurs tests neuropsychologiques ont été complétés par 115 adultes sains âgés de 45 à 87 ans. Une analyse en composantes principales réalisée sur l'ensemble des questions du QAM a permis d’identifier sept grands domaines de plainte. Des analyses subséquentes ont révélé que les plaintes les plus fréquemment rapportées par les participants sont associées à des situations où des facteurs internes et externes interfèrent avec la performance mnésique. Les analyses ont aussi montré que les plaintes relatives à des oublis dont les conséquences menacent l’autonomie et la sécurité témoigneraient de problèmes cognitifs et fonctionnels plus sévères. Enfin, nos résultats ont indiqué que les différents domaines de plainte reflètent globalement les problèmes cognitifs objectifs. Aucune association n’a été trouvée entre la plainte et la plupart des caractéristiques démographiques. La seconde étude (Chapitre III) avait pour but de caractériser la plainte cognitive dans le TCL ainsi que son évolution dans la progression de la démence. L’étude cherchait aussi à déterminer si les changements dans certains domaines de plainte étaient reliés au déclin de ii fonctions cognitives spécifiques chez les individus avec TCL qui ont progressé vers la démence (progresseurs). Des personnes avec TCL et des individus âgés sains ont été évalués annuellement pendant trois ans. Le QAM et le Multifactorial Memory Questionnaire (MMQ; Fort, Holl, Kaddour, & Gana, 2004) ont été utilisés pour mesurer leurs plaintes. Les résultats ont révélé que les progresseurs rapportaient davantage de difficultés associées à la mémorisation de contenus complexes (ex. : textes ou conversations), d’événements récents et d’informations sur leurs proches que les personnes âgées saines et ce, jusqu’à trois ans avec le diagnostic de démence. L’analyse des effets de groupe a indiqué que l’intensité des plaintes des progresseurs semble être demeurée stable durant le suivi. Il est donc possible qu’une proportion des progresseurs présentent une méconnaissance de leurs difficultés cognitives. Cependant, des analyses corrélationnelles ont montré que l’augmentation des plaintes reliées à trois domaines était associée à l’accroissement de certaines atteintes cognitives durant les trois ans. Ainsi, certaines plaintes pourraient permettre de mieux comprendre les difficultés cognitives qui sont vécues par la personne avec TCL. Les implications théoriques et cliniques de ces résultats seront discutées dans le dernier chapitre de la thèse (Chapitre IV).
Resumo:
Adolescent idiopathic scoliosis (AIS) is a deformity of the spine manifested by asymmetry and deformities of the external surface of the trunk. Classification of scoliosis deformities according to curve type is used to plan management of scoliosis patients. Currently, scoliosis curve type is determined based on X-ray exam. However, cumulative exposure to X-rays radiation significantly increases the risk for certain cancer. In this paper, we propose a robust system that can classify the scoliosis curve type from non invasive acquisition of 3D trunk surface of the patients. The 3D image of the trunk is divided into patches and local geometric descriptors characterizing the surface of the back are computed from each patch and forming the features. We perform the reduction of the dimensionality by using Principal Component Analysis and 53 components were retained. In this work a multi-class classifier is built with Least-squares support vector machine (LS-SVM) which is a kernel classifier. For this study, a new kernel was designed in order to achieve a robust classifier in comparison with polynomial and Gaussian kernel. The proposed system was validated using data of 103 patients with different scoliosis curve types diagnosed and classified by an orthopedic surgeon from the X-ray images. The average rate of successful classification was 93.3% with a better rate of prediction for the major thoracic and lumbar/thoracolumbar types.
Resumo:
Objective To determine scoliosis curve types using non invasive surface acquisition, without prior knowledge from X-ray data. Methods Classification of scoliosis deformities according to curve type is used in the clinical management of scoliotic patients. In this work, we propose a robust system that can determine the scoliosis curve type from non invasive acquisition of the 3D back surface of the patients. The 3D image of the surface of the trunk is divided into patches and local geometric descriptors characterizing the back surface are computed from each patch and constitute the features. We reduce the dimensionality by using principal component analysis and retain 53 components using an overlap criterion combined with the total variance in the observed variables. In this work, a multi-class classifier is built with least-squares support vector machines (LS-SVM). The original LS-SVM formulation was modified by weighting the positive and negative samples differently and a new kernel was designed in order to achieve a robust classifier. The proposed system is validated using data from 165 patients with different scoliosis curve types. The results of our non invasive classification were compared with those obtained by an expert using X-ray images. Results The average rate of successful classification was computed using a leave-one-out cross-validation procedure. The overall accuracy of the system was 95%. As for the correct classification rates per class, we obtained 96%, 84% and 97% for the thoracic, double major and lumbar/thoracolumbar curve types, respectively. Conclusion This study shows that it is possible to find a relationship between the internal deformity and the back surface deformity in scoliosis with machine learning methods. The proposed system uses non invasive surface acquisition, which is safe for the patient as it involves no radiation. Also, the design of a specific kernel improved classification performance.
Resumo:
The thesis report results obtained from a detailed analysis of the fluctuations of the rheological parameters viz. shear and normal stresses, simulated by means of the Stokesian Dynamics method, of a macroscopically homogeneous sheared suspension of neutrally buoyant non-Brownian suspension of identical spheres in the Couette gap between two parallel walls in the limit of vanishingly small Reynolds numbers using the tools of non-linear dynamics and chaos theory for a range of particle concentration and Couette gaps. The thesis used the tools of nonlinear dynamics and chaos theory viz. average mutual information, space-time separation plots, visual recurrence analysis, principal component analysis, false nearest-neighbor technique, correlation integrals, computation of Lyapunov exponents for a range of area fraction of particles and for different Couette gaps. The thesis observed that one stress component can be predicted using another stress component at the same area fraction. This implies a type of synchronization of one stress component with another stress component. This finding suggests us to further analysis of the synchronization of stress components with another stress component at the same or different area fraction of particles. The different model equations of stress components for different area fraction of particles hints at the possible existence a general formula for stress fluctuations with area fraction of particle as a parameter
Resumo:
The present study focused on the quality of rainwater at various land use locations and its variations on interaction with various domestic rainwater harvesting systems.Sampling sites were selected based upon the land use pattern of the locations and were classified as rural, urban, industrial and sub urban. Rainwater samples were collected from the south west monsoon of May 2007 to north east monsoon of October 2008, from four sampling sites namely Kothamangalam, Emakulam, Eloor and Kalamassery, in Ernakulam district of the State of Kerala, which characterized typical rural, urban, industrial and suburban locations respectively. Rain water samples at various stages of harvesting were also collected. The samples were analyzed according to standard procedures and their physico-chemical and microbiological parameters were determined. The variations of the chemical composition of the rainwater collected were studied using statistical methods. It was observed that 17.5%, 30%, 45.8% and 12.1% of rainwater samples collected at rural, urban, industrial and suburban locations respectively had pH less than 5.6, which is considered as the pH of cloud water at equilibrium with atmospheric CO,.Nearly 46% of the rainwater samples were in acidic range in the industrial location while it was only 17% in the rural location. Multivariate statistical analysls was done using Principal Component Analysis, and the sources that inf1uence the composition of rainwater at each locations were identified .which clearly indicated that the quality of rain water is site specific and represents the atmospheric characteristics of the free fall The quality of harvested rainwater showed significant variations at different stages of harvesting due to deposition of dust from the roof catchment surface, leaching of cement constituents etc. Except the micro biological quality, the harvested rainwater satisfied the Indian Standard guide lines for drinking water. Studies conducted on the leaching of cement constituents in water concluded that tanks made with ordinary portland cement and portland pozzolana cement could be safely used for storage of rain water.
Resumo:
Any automatically measurable, robust and distinctive physical characteristic or personal trait that can be used to identify an individual or verify the claimed identity of an individual, referred to as biometrics, has gained significant interest in the wake of heightened concerns about security and rapid advancements in networking, communication and mobility. Multimodal biometrics is expected to be ultra-secure and reliable, due to the presence of multiple and independent—verification clues. In this study, a multimodal biometric system utilising audio and facial signatures has been implemented and error analysis has been carried out. A total of one thousand face images and 250 sound tracks of 50 users are used for training the proposed system. To account for the attempts of the unregistered signatures data of 25 new users are tested. The short term spectral features were extracted from the sound data and Vector Quantization was done using K-means algorithm. Face images are identified based on Eigen face approach using Principal Component Analysis. The success rate of multimodal system using speech and face is higher when compared to individual unimodal recognition systems
Resumo:
The composition and variability of heterotrophic bacteria along the shelf sediments of south west coast of India and its relationship with the sediment biogeochemistry was investigated. The bacterial abundance ranged from 1.12 x 103 – 1.88 x 106 CFU g-1 dry wt. of sediment. The population showed significant positive correlation with silt (r = 0.529, p< 0.05), organic carbon (OC) (r = 0.679, p< 0.05), total nitrogen (TN) (r = 0.638, p< 0.05), total protein (TPRT) (r = 0.615, p< 0.05) and total carbohydrate (TCHO) (r = 0.675, p< 0.05) and significant negative correlation with sand (r = -0.488, p< 0.05). Community was mainly composed of Bacillus, Alteromonas, Vibrio, Coryneforms, Micrococcus, Planococcus, Staphylococcus, Moraxella, Alcaligenes, Enterobacteriaceae, Pseudomonas, Acinetobacter, Flavobacterium and Aeromonas. BIOENV analysis explained the best possible environmental parameters i.e., carbohydrate, total nitrogen, temperature, pH and sand at 50m depth and organic matter, BPC, protein, lipid and temperature at 200m depth controlling the distribution pattern of heterotrophic bacterial population in shelf sediments. The Principal Component Analysis (PCA) of the environmental variables showed that the first and second principal component accounted for 65% and 30.6% of the data variance respectively. Canonical Correspondence Analysis (CCA) revealed a strong correspondence between bacterial distribution and environmental variables in the study area. Moreover, non-metric MDS (Multidimensional Scaling) analysis demarcated the northern and southern latitudes of the study area based on the bioavailable organic matter
Resumo:
In this paper, we propose a handwritten character recognition system for Malayalam language. The feature extraction phase consists of gradient and curvature calculation and dimensionality reduction using Principal Component Analysis. Directional information from the arc tangent of gradient is used as gradient feature. Strength of gradient in curvature direction is used as the curvature feature. The proposed system uses a combination of gradient and curvature feature in reduced dimension as the feature vector. For classification, discriminative power of Support Vector Machine (SVM) is evaluated. The results reveal that SVM with Radial Basis Function (RBF) kernel yield the best performance with 96.28% and 97.96% of accuracy in two different datasets. This is the highest accuracy ever reported on these datasets
Resumo:
Learning Disability (LD) is a classification including several disorders in which a child has difficulty in learning in a typical manner, usually caused by an unknown factor or factors. LD affects about 15% of children enrolled in schools. The prediction of learning disability is a complicated task since the identification of LD from diverse features or signs is a complicated problem. There is no cure for learning disabilities and they are life-long. The problems of children with specific learning disabilities have been a cause of concern to parents and teachers for some time. The aim of this paper is to develop a new algorithm for imputing missing values and to determine the significance of the missing value imputation method and dimensionality reduction method in the performance of fuzzy and neuro fuzzy classifiers with specific emphasis on prediction of learning disabilities in school age children. In the basic assessment method for prediction of LD, checklists are generally used and the data cases thus collected fully depends on the mood of children and may have also contain redundant as well as missing values. Therefore, in this study, we are proposing a new algorithm, viz. the correlation based new algorithm for imputing the missing values and Principal Component Analysis (PCA) for reducing the irrelevant attributes. After the study, it is found that, the preprocessing methods applied by us improves the quality of data and thereby increases the accuracy of the classifiers. The system is implemented in Math works Software Mat Lab 7.10. The results obtained from this study have illustrated that the developed missing value imputation method is very good contribution in prediction system and is capable of improving the performance of a classifier.
Resumo:
Learning Disability (LD) is a neurological condition that affects a child’s brain and impairs his ability to carry out one or many specific tasks. LD affects about 15 % of children enrolled in schools. The prediction of LD is a vital and intricate job. The aim of this paper is to design an effective and powerful tool, using the two intelligent methods viz., Artificial Neural Network and Adaptive Neuro-Fuzzy Inference System, for measuring the percentage of LD that affected in school-age children. In this study, we are proposing some soft computing methods in data preprocessing for improving the accuracy of the tool as well as the classifier. The data preprocessing is performed through Principal Component Analysis for attribute reduction and closest fit algorithm is used for imputing missing values. The main idea in developing the LD prediction tool is not only to predict the LD present in children but also to measure its percentage along with its class like low or minor or major. The system is implemented in Mathworks Software MatLab 7.10. The results obtained from this study have illustrated that the designed prediction system or tool is capable of measuring the LD effectively
Resumo:
In this paper an attempt has been made to determine the number of Premature Ventricular Contraction (PVC) cycles accurately from a given Electrocardiogram (ECG) using a wavelet constructed from multiple Gaussian functions. It is difficult to assess the ECGs of patients who are continuously monitored over a long period of time. Hence the proposed method of classification will be helpful to doctors to determine the severity of PVC in a patient. Principal Component Analysis (PCA) and a simple classifier have been used in addition to the specially developed wavelet transform. The proposed wavelet has been designed using multiple Gaussian functions which when summed up looks similar to that of a normal ECG. The number of Gaussians used depends on the number of peaks present in a normal ECG. The developed wavelet satisfied all the properties of a traditional continuous wavelet. The new wavelet was optimized using genetic algorithm (GA). ECG records from Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) database have been used for validation. Out of the 8694 ECG cycles used for evaluation, the classification algorithm responded with an accuracy of 97.77%. In order to compare the performance of the new wavelet, classification was also performed using the standard wavelets like morlet, meyer, bior3.9, db5, db3, sym3 and haar. The new wavelet outperforms the rest
Resumo:
Phosphorus fractionation was employed to find the bioavailability of phosphorus and its seasonal variations in the Panangad region of Cochin estuary, the largest estuarine system in the southwest coast of India. Sequential extraction of the surficial sediments using chelating agents was taken as a tool for this. Phosphate in the water column showed seasonal variations, with high values during the monsoon months, suggesting external runoff. Sediment texture was found to be the main factor influencing the spatial distribution of the geochemical parameters in the study region. Similarly, total phosphorus also showed granulometric dependence and it ranged between 319.54 and 2,938.83 μg/g. Calcium-bound fraction was the main phosphorus pool in the estuary. Significant spatial variations were observed for all bioavailable fractions; iron-bound inorganic phosphorus (5.04–474.24 μg/g), calcium-bound inorganic phosphorus (11.16–826.09 μg/g), and acidsoluble organic phosphorus (22.22–365.86 μg/g). Among the non-bioavailable phosphorus, alkalisoluble organic fraction was the major one (51.92– 1,002.45 μg/g). Residual organic phosphorus was K. R. Renjith (B) · N. Chandramohanakumar · M. M. Joseph Department of Chemical Oceanography, School of Marine Sciences, Cochin University of Science and Technology, Kochi 682016, Kerala, India e-mail: renjithaqua@gmail.com comparatively smaller fraction (3.25–14.64% of total). The sandy and muddy stations showed distinct fractional composition and the speciation study could endorse the overall geochemical character. There could be buffering of phosphorus, suggested by the increase in the percentage of bioavailable fractions during the lean premonsoon period, counteracting the decreases in the external loads. Principal component analysis was employed to find the possible processes influencing the speciation of phosphorus in the study region
Resumo:
Concentrations and distributions of trace metals (Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, and Zn) in surficial sediments of the Cochin backwaters were studied during both monsoon and pre-monsoon periods. Spatial variations were in accordance with textural charaterstics and organic matter content. A principal component analysis distinguished three zones with different metal accumulation capacity: (i) highest levels in north estuary, (ii) moderate levels in central zone, and (iii) lowest levels in southern part. Trace metal enrichments are mainly due to anthropogenic contribution of industrial, domestic, and agricultural effluents, whose effect is enhanced by settling of metals due to organic flocculation and inorganic precipitation associated with salinity changes. Enrichments factors using Fe as a normalizer showed that metal contamination was the product of anthropogenic activities. An assessment of degree of pollution-categorized sediments as moderately polluted with Cu and Pb, moderately-to-heavily polluted with Zn, and heavily-to-extremely polluted with Cd. Concentrations at many sites largely exceed NOAA ERL (e.g., Cu, Cr, and Pb) or ERM (e.g., Cd, Ni, and Zn). This means that adverse effects for benthic organisms are possible or even highly probable.
Resumo:
This article present the result from a study of two sediment cores collected from the environmentally distinct zones of CES. Accumulation status of five toxic metals: Cadmium (Cd), Chromium (Cr), Cobalt (Co), Copper (Cu) and Lead (Pb) were analyzed. Besides texture and CHNS were determined to understand the composition of the sediment. Enrichment Factor (EF) and Anthropogenic Factor (AF) were used to differentiate the typical metal sources. Metal enrichment in the cores revealed heavy load at the northern (NS1 ) region compared with the southern zone (SS1). Elevation of metal content in core NS1 showed the industrial input. Statistical analyses were employed to understand the origin of metals in the sediment samples. Principal Component Analysis (PCA) distinguishes the two zones with different metal accumulation capacity: highest at NS1 and lowest at SS1. Correlation analysis revealed positive significant relation only in core NS1, adhering to the exposition of the intensified industrial pollution