- About
-
The Global ETD Search service is a free service for researchers
to find electronic theses and dissertations. This service is
provided by the
Networked Digital Library of Theses and Dissertations.
Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
Search results
Showing 61 to 70 of 70 (0.027 seconds)Spelling suggestions: "subject:"[een] FUZZY C-MEANS"" "subject:"[enn] FUZZY C-MEANS""
| 61 |
Apport de la morphologie tridimensionnelle de la colonne vertébrale et du bassin à la scoliose idiopathique de l’adolescenceDubé, Evelyne 08 1900 (has links)
La scoliose idiopathique de l’adolescence (SIA) est une déformation de la colonne vertébrale d’origine inconnue. Bien qu’elle soit encore aujourd’hui mesurée et classifiée en utilisant des radiographies bidimensionnelles (2D), il est largement rapporté dans la littérature qu’il s’agit d’une affection dans les trois plans de l’espace. Aussi, il semble que le bassin soit impliqué dans la déformation scoliotique, puisqu’il constitue l’assise de la colonne vertébrale et que son orientation influe sur l’équilibre postural. Ainsi, l'asymétrie du bassin et son attitude posturale pourraient être des mécanismes compensatoires de la scoliose idiopathique ou encore être les agents qui déclenchent la déformation du rachis. L’objectif de ce travail est de déterminer la relation entre la morphologie tridimensionnelle (3D) de la colonne vertébrale et du bassin et les déformations scoliotiques groupées selon la méthode de Lenke et de connaître les liens entre ces paramètres morphologiques et l’angle de Cobb.
Pour ce faire, 80 filles atteintes de la SIA ont participé à l’étude. Plus précisément, 32 sujets étaient atteintes d’une scoliose thoracique, 23 d’une scoliose thoraco-lombaire et 25 d’une scoliose lombaire. Des radiographies simultanées des plans postéro-antérieur et latéral en position debout ont été prises au moyen du système EOS. Quinze repères anatomiques sur chacune des vertèbres entre T1 à L5 et vingt-et-un sur le bassin ont été identifiés sur les paires de radiographies. La reconstruction tridimensionnelle de la colonne vertébrale et du bassin a été faite à partir des repères anatomiques. Au total, cinq paramètres sur la colonne vertébrale et trois sur le bassin ont été calculés afin d’identifier la morphologie des déformations scoliotiques thoraciques, thoraco-lombaires et lombaires. L’algorithme de classification non-supervisée de la logique floue ou fuzzy c-means (FCM) a été utilisé pour classifier les sujets. Des classifications à deux et trois classes ont été faites avec les données normalisées et non-normalisées, c’est-à-dire en faisant ou en ne faisant pas abstraction au niveau de la courbure scoliotique. Des analyses de variances à un facteur (ANOVA) avec post-hoc ont été menées sur les classifications à deux groupes et deux classes, alors que des analyses multivariées (MANOVA) avec post-hoc ont été réalisées sur les classifications à trois groupes et trois classes non-normalisés et normalisés.
L’angle de Cobb du segment thoracique principal est significativement différent pour les trois types de scolioses. Cependant, ces différences pourraient être associées aux segments analysés et à la sévérité de la courbure. Avec les données normalisées, les scolioses thoraciques L1 se regroupent ensemble pour la classification à deux classes et se divisent en deux pour la classification à trois classes. Les paramètres de la cyphose (p = 0,000), de la lordose (p = 0,000) et de l’orientation du plan de courbure maximale (PCM) (p = 0,000) sont ceux qui divisent ces sujets. Quant aux L5 et L6, peu importe la classification, ils se rassemblent généralement dans une même classe. Aussi, des corrélations de Pearson ont été réalisées en fonction de l’angle de Cobb, afin de déceler des liens entre les types de déformation et les paramètres morphologiques. Le bassin ne semble pas avoir d’impact sur l’issu des classifications, mais il est corrélé avec les déviations scoliotique des sujets lombaires. En effet, la version pelvienne (r = -0,433; p = 0,031) est en relation inverse, tandis que la pente sacrée est en relation directe (r = 0,419; p = 0,037).
En résumé, les résultats de cette étude indiquent que l’apport de la morphologie 3D de la colonne vertébrale et du bassin aux déformations scoliotiques thoraciques (L1), thoraco-lombaires (L5) et lombaires (L6) apporte des informations cliniquement pertinentes. Nos résultats obtenus par logique floue concernant les sujets thoraciques appuient ceux de la littérature, à savoir que ces sujets ne sont pas tous hypocyphosés. De nouveaux paramètres, tels que la lordose et l’orientation du PCM, viennent renforcer l’idée qu’il existe des sous-groupes parmi les scoliotiques thoraciques. Finalement, bien que d’un point de vue visuel, les sujets thoraco-lombaires et lombaires soient différents, du côté de la morphologie tridimensionnelle, ces sujets sont inséparables. / Adolescent idiopathic scoliosis (AIS) is a deviation of the spine of unknown origin. Although it is still classified and measured using two-dimensional X-ray, it is widely reported in the literature that scoliosis is a deviation in the three planes of space. Also, it seems that the pelvis is involved in the scoliosis deformity, as it is the foundation of the spine and its orientation influence on postural balance. Thus, the asymmetry of the pelvis and postural attitude could be compensatory mechanisms of idiopathic scoliosis or be agents that trigger this disease. The aim of this work is to determine the relationship between the three-dimensional morphology of the spine and pelvis of scoliosis grouped according to Lenke’s classification and to know the links between morphological parameters and the Cobb angle.
Eighty girls with the SIA participated in this study. Specifically, 32 subjects were suffering from thoracic scoliosis, 23 thoracolumbar scoliosis and 25 lumbar scoliosis. Simultaneous radiographs of posterior-anterior and lateral planes in standing position were taken with the EOS system. Fifteen anatomical landmarks on each of the vertebrae between T1 and L5 on the spine and twenty-one on the pelvis, have been identified on the pairs of radiographs. The three-dimensional reconstruction of the spine and pelvis was made from the two radiographs and the anatomic landmarks. A total of five parameters on the spine and three on the pelvis were calculated to identify the morphology of thoracic, thoracolumbar and lumbar deformations. The unsupervised classification algorithm of fuzzy c-means (FCM) was used to classify subjects. Classifications with two and three classes were made with non-standardized and normalized data, i.e. by omitting the level of the scoliosis curve. Analysis of variance (ANOVA) with post-hoc was conducted on classifications with two groups and classes and multivariate analysis (MANOVA) with post-hoc were made on classifications with three groups and classes.
The Cobb angle of the main thoracic segment was significantly different for the three types of scoliosis. However, these differences might be attributed to the analyzed segment and severity of the curve. With the normalized data, the thoracic scoliosis L1 subjects regroup in classification with two classes and divides into two classes in the classification with three classes. These classes are divided according to the kyphosis (p = 0.000), lordosis (p = 0.000) and the orientation of plane of maximum curvature (PMC) (p = 0.000) parameters. Regardless of classification, L5 and L6 usually gather in the same class. Also, Pearson correlations were made according to the Cobb angle, in order to detect the relationship between the types of deformation and morphological parameters. Finally, the pelvis had no impact on classifications, but it is correlated with the scoliotic deviation of the lumbar scoliosis. The pelvic tilt (r = -0.433; p = 0.031) is inversely correlated, while the sacral slope has a direct relationship (r = 0.419; p = 0.037).
In conclusion, the results of this study indicate that the contribution of the 3D morphology of the spine and pelvis to the thoracic (L1), thoracolumbar (L5) and lumbar (L6) scoliosis provides clinically relevant information. Our results gained by fuzzy-c-means support those in the literature, according to which these subjects are not all hypokyphotic. New parameters such as lordosis and orientation of the PMC, reinforce the idea that there are subgroups within the thoracic scoliosis. Finally, although the thoracolumbar and lumbar subjects appear to differ from a visual standpoint, these subjects are inseparable according to the three-dimensional morphology parameters.
|
| 62 |
Driving data pattern recognition for intelligent energy management of plug-in hybrid electric vehiclesMunthikodu, Sreejith 19 August 2019 (has links)
This work focuses on the development and testing of new driving data pattern recognition intelligent system techniques to support driver adaptive, real-time optimal power control and energy management of hybrid electric vehicles (HEVs) and plug-in hybrid electric vehicles (PHEVs). A novel, intelligent energy management approach that combines vehicle operation data acquisition, driving data clustering and pattern recognition, cluster prototype based power control and energy optimization, and real-time driving pattern recognition and optimal energy management has been introduced. The method integrates advanced machine learning techniques and global optimization methods form the driver adaptive optimal power control and energy management. Fuzzy C-Means clustering algorithm is used to identify the representative vehicle operation patterns from collected driving data. Dynamic Programming (DA) based off-line optimization is conducted to obtain the optimal control parameters for each of the identified driving patterns. Artificial Neural Networks (ANN) are trained to associate each of the identified operation patterns with the optimal energy management plan to support real-time optimal control. Implementation and advantages of the new method are demonstrated using the 2012 California household travel survey data, and driver-specific data collected from the city of Victoria, BC Canada. / Graduate
|
| 63 |
Frequency Analysis of Droughts Using Stochastic and Soft Computing TechniquesSadri, Sara January 2010 (has links)
In the Canadian Prairies recurring droughts are one of the realities which can
have significant economical, environmental, and social impacts. For example,
droughts in 1997 and 2001 cost over $100 million on different sectors. Drought frequency
analysis is a technique for analyzing how frequently a drought event of a given
magnitude may be expected to occur. In this study the state of the science related
to frequency analysis of droughts is reviewed and studied. The main contributions
of this thesis include development of a model in Matlab which uses the qualities of
Fuzzy C-Means (FCMs) clustering and corrects the formed regions to meet the criteria
of effective hydrological regions. In FCM each site has a degree of membership in
each of the clusters. The algorithm developed is flexible to get number of regions and
return period as inputs and show the final corrected clusters as output for most case
scenarios. While drought is considered a bivariate phenomena with two statistical
variables of duration and severity to be analyzed simultaneously, an important step
in this study is increasing the complexity of the initial model in Matlab to correct
regions based on L-comoments statistics (as apposed to L-moments). Implementing
a reasonably straightforward approach for bivariate drought frequency analysis using
bivariate L-comoments and copula is another contribution of this study. Quantile estimation at ungauged sites for return periods of interest is studied by introducing two
new classes of neural network and machine learning: Radial Basis Function (RBF)
and Support Vector Machine Regression (SVM-R). These two techniques are selected
based on their good reviews in literature in function estimation and nonparametric
regression. The functionalities of RBF and SVM-R are compared with traditional
nonlinear regression (NLR) method. As well, a nonlinear regression with regionalization
method in which catchments are first regionalized using FCMs is applied and
its results are compared with the other three models. Drought data from 36 natural
catchments in the Canadian Prairies are used in this study. This study provides a
methodology for bivariate drought frequency analysis that can be practiced in any
part of the world.
|
| 64 |
Decision Making System Algorithm On Menopause Data SetBacak, Hikmet Ozge 01 September 2007 (has links) (PDF)
Multiple-centered clustering method and decision making system algorithm on menopause data set depending on multiple-centered clustering are described in this study. This method consists of two stages. At the first stage, fuzzy C-means (FCM) clustering algorithm is applied on the data set under consideration with a high number of cluster centers. As the output of FCM, cluster centers and membership function values for each data member is calculated. At the second stage, original cluster centers obtained in the first stage are merged till the new numbers of clusters are reached. Merging process relies upon a &ldquo / similarity measure&rdquo / between clusters defined in the thesis. During the merging process, the cluster center coordinates do not change but the data members in these clusters are merged in a new cluster. As the output of this method, therefore, one obtains clusters which include many cluster centers.
In the final part of this study, an application of the clustering algorithms &ndash / including the multiple centered clustering method &ndash / a decision making system is constructed using a special data on menopause treatment. The decisions are based on the clusterings created by the algorithms already discussed in the previous chapters of the thesis. A verification of the decision making system /
v
decision aid system is done by a team of experts from the Department of Department of Obstetrics and Gynecology of Hacettepe University under the guidance of Prof. Sinan Beksaç / .
|
| 65 |
Frequency Analysis of Droughts Using Stochastic and Soft Computing TechniquesSadri, Sara January 2010 (has links)
In the Canadian Prairies recurring droughts are one of the realities which can
have significant economical, environmental, and social impacts. For example,
droughts in 1997 and 2001 cost over $100 million on different sectors. Drought frequency
analysis is a technique for analyzing how frequently a drought event of a given
magnitude may be expected to occur. In this study the state of the science related
to frequency analysis of droughts is reviewed and studied. The main contributions
of this thesis include development of a model in Matlab which uses the qualities of
Fuzzy C-Means (FCMs) clustering and corrects the formed regions to meet the criteria
of effective hydrological regions. In FCM each site has a degree of membership in
each of the clusters. The algorithm developed is flexible to get number of regions and
return period as inputs and show the final corrected clusters as output for most case
scenarios. While drought is considered a bivariate phenomena with two statistical
variables of duration and severity to be analyzed simultaneously, an important step
in this study is increasing the complexity of the initial model in Matlab to correct
regions based on L-comoments statistics (as apposed to L-moments). Implementing
a reasonably straightforward approach for bivariate drought frequency analysis using
bivariate L-comoments and copula is another contribution of this study. Quantile estimation at ungauged sites for return periods of interest is studied by introducing two
new classes of neural network and machine learning: Radial Basis Function (RBF)
and Support Vector Machine Regression (SVM-R). These two techniques are selected
based on their good reviews in literature in function estimation and nonparametric
regression. The functionalities of RBF and SVM-R are compared with traditional
nonlinear regression (NLR) method. As well, a nonlinear regression with regionalization
method in which catchments are first regionalized using FCMs is applied and
its results are compared with the other three models. Drought data from 36 natural
catchments in the Canadian Prairies are used in this study. This study provides a
methodology for bivariate drought frequency analysis that can be practiced in any
part of the world.
|
| 66 |
Regionalization Of Hydrometeorological Variables In India Using Cluster AnalysisBharath, R 09 1900 (has links) (PDF)
Regionalization of hydrometeorological variables such as rainfall and temperature is necessary for various applications related to water resources planning and management. Sampling variability and randomness associated with the variables, as well as non-availability and paucity of data pose a challenge in modelling the variables. This challenge can be addressed by using stochastic models that utilize information from hydrometeorologically similar locations for modelling the variables. A set of locations that are hydrometeorologically similar are referred to as homogeneous region or pooling group and the process of identifying a homogeneous region is referred to as regionalization. The thesis concerns development of new approaches to regionalization of (i) extreme rainfall,(ii) maximum and minimum temperatures, and (iii) rainfall together with maximum and minimum temperatures.
Regionalization of extreme rainfall and frequency analysis based on resulting regions yields quantile estimates that find use in design of water control (e.g., barrages, dams, levees) and conveyance structures (e.g., culverts, storm sewers, spillways) to mitigate damages that are likely due to floods triggered by extreme rainfall, and land-use planning and management. Regionalization based on both rainfall and temperature yield regions that could be used to address a wide spectrum of problems such as meteorological drought analysis, agricultural planning to cope with water shortages during droughts, downscaling of precipitation and temperature.
Conventional approaches to regionalization of extreme rainfall are based extensively on statistics derived from extreme rainfall. Therefore delineated regions are susceptible to sampling variability and randomness associated with extreme rainfall records, which is undesirable. To address this, the idea of forming regions by considering attributes for regionalization as seasonality measure and site location indicators (which could be determined even for ungauged locations) is explored. For regionalization, Global Fuzzy c-means (GFCM) cluster analysis based methodology is developed in L-moment framework. The methodology is used to arrive at a set of 25 homogeneous extreme rainfall regions over India considering gridded rainfall records at daily scale, as there is dearth of regionalization studies on extreme rainfall in India Results are compared with those based on commonly used region of influence (ROI) approach that forms site-specific regions for quantile estimation, but lacks ability to delineate a geographical area into a reasonable number of homogeneous regions. Gridded data constitute spatially averaged rainfall that might originate from a different process (more synoptic) than point rainfall (more convective). Therefore to investigate utility of the developed GFCM methodology in arriving at meaningful regions when applied to point rainfall data, the methodology is applied to daily rainfall records available for 1032 gauges in Karnataka state of India. The application yielded 22 homogeneous extreme rainfall regions. Experiments carried out to examine utility of GFCM and ROI based regions in arriving at quantile estimates for ungauged sites in the study area reveal that performance of GFCM methodology is fairly close to that of ROI approach. Errors were marginally lower in the case of GFCM approach in analysis with observed point rainfall data over Karnataka, while its converse was noted in the case of analysis with gridded rainfall data over India. Neither of the approaches (CA, ROI) was found to be consistent in yielding least error in quantile estimates over all the sites.
The existing approaches to regionalization of temperature are based on temperature time series or their related statistics, rather than attributes effecting temperature in the study area. Therefore independent validation of the delineated regions for homogeneity in temperature is not possible. Another drawback of the existing approaches is that they require adequate number of sites with contemporaneous temperature records for regionalization, because the delineated regions are susceptible to sampling variability and randomness associated with the temperature records that are often (i) short in length, (ii) limited over contemporaneous time period and (iii) spatially sparse. To address these issues, a two-stage clustering approach is developed to arrive at regions that are homogeneous in terms of both monthly maximum and minimum temperatures ( and ). First-stage of the approach involves (i) identifying a common set of possible predictors (LSAVs) influencing and over the entire study area, and (ii) using correlations of those predictors with and along with location indicators (latitude, longitude and altitude) as the basis to delineate sites in the study area into hard clusters through global k-means clustering algorithm. The second stage involves (i) identifying appropriate LSAVs corresponding to each of the first-stage clusters, which could be considered as potential predictors, and (ii) using the potential predictors along with location indicators (latitude, longitude and altitude) as the basis to partition each of the first-stage clusters into homogeneous temperature regions through global fuzzy c-means clustering algorithm. A set of 28 homogeneous temperature regions was delineated over India using the proposed approach. Those regions are shown to be effective when compared to an existing set of 6 temperature regions over India for which inter-site cross-correlations were found to be weak and negative for several months, which is undesirable. Effectiveness of the newly formed regions is demonstrated. Utility of the proposed maxTminT
homogeneous temperature regions in arriving at PET estimates for ungauged locations within the study area was demonstrated. The estimates were found to be better when compared to those based on the existing regions.
The existing approaches to regionalization of hydrometeorological variables are based on principal components (PCs)/ statistics/indices determined from time-series of those variables at monthly and seasonal scale. An issue with use of PCs for regionalization is that they have to be extracted from contemporaneous records of hydrometeorological variables. Therefore delineated regions may not be effective when the available records are limited over contemporaneous time period. A drawback associated with the use of statistics/indices is that they (i) may not be meaningful when data exhibit nonstationarity and (ii) do not encompass complete information in the original time series. Consequently the resulting regions may not be effective for the desired purpose. To address these issues, a new approach is proposed. It considers information extracted from wavelet transformations of the observed multivariate hydrometeorological time series as the basis for regionalization by global fuzzy c-means clustering procedure. The approach can account for dynamic variability in the time series and its nonstationarity (if any). Effectiveness of the proposed approach in forming homogeneous hydrometeorological regions is demonstrated by application to India, as there are no prior attempts to form such regions over the country. The investigations resulted in identification of 29 regions over India, which are found to be effective and meaningful. Drought Severity-Area-Frequency (SAF) curves are developed for each of the newly formed regions considering the drought index to be Standardized Precipitation Evapotranspiration Index (SPEI).
|
| 67 |
Проектирование системы для отбора заявок на финансирование цифровых проектов в регионе : магистерская диссертация / Design of a system for selecting applications for funding digital projects in the regionАлимпиев, Я. О., Alimpiev, Y. O. January 2024 (has links)
The object of this study is the process of selecting applications for funding digital projects in the region. The subject of the study is a design system that includes algorithms and methods used to select applications for funding digital projects in the region. The purpose of this work is to develop an approach to selecting digital projects for funding that takes into account regional characteristics and the efficiency of resource use. This work studies the market for digital projects / as well as methods and algorithms for their selection. Several algorithms were compared to determine the most suitable for selecting digital projects. The predictive ability of the selected algorithm was assessed. A structural diagram of the algorithm is presented and an assessment of the economic and product efficiency of the system designed on the basis of this algorithm is made. Research methods - literature review / statistical and economic analysis / fuzzy clustering. The result of the work is a system for selecting applications for funding digital projects. / Объектом данного исследования является процесс отбора заявок на финансирование цифровых проектов в регионе. Предметом исследования является система проектирования, включающая алгоритмы и методики, используемые для отбора заявок на финансирование цифровых проектов в регионе. Цель данной работы – разработка подхода к отбору цифровых проектов для финансирования, учитывающего региональные особенности и эффективность использования ресурсов. В данной работе изучается рынок цифровых проектов, а также методы и алгоритмы их отбора. Проведено сравнение нескольких алгоритмов с целью определения, наиболее подходящего для отбора цифровых проектов. Оценена прогнозирующая способность выбранного алгоритма. Представлена структурная схема алгоритма и выполнена оценка экономической и продуктовой эффективности системы, спроектированной на основе данного алгоритма. Методы исследования – литературный обзор, статистико-экономический анализ, нечеткая кластеризация. Результатом работы является система для отбора заявок на финансирование цифровых проектов.
|
| 68 |
Design of an electrocardiographic lead reconstruction algorithm using machine learning in the context of ambulatory monitoringGrande Fidalgo, Alejandro 16 January 2025 (has links)
[ES] Esta tesis doctoral presenta un algoritmo para reconstruir el registro electrocardiográfico (ECG) estándar del sistema de 12 derivaciones utilizando un sistema reducido de derivaciones independientes mediante el uso de modelos de aprendizaje automático, centrándose en su integración en un sistema de monitorización ambulatoria. Los métodos tradicionales de reconstrucción de ECG se basan en enfoques basados en combinaciones lineales, con una exploración limitada de los métodos de evaluación y de las posiciones de los electrodos. Esta tesis evalúa la eficacia de nuevas redes neuronales artificiales y algoritmos basados en fuzzy c-means en comparación con los métodos clásicos de regresión lineal, destacando un rendimiento superior y subrayando la importancia de la explicabilidad del modelo. Se exploran otras mejoras, como comités de expertos y modelos difusos, para aumentar la precisión y la eficacia. La validación clínica realizada en el Hospital Clínico Universitario de València y en el Hospital General Universitario de València demuestran la eficacia del algoritmo en la reconstrucción precisa de derivaciones, facilitando el camino para aplicaciones de monitorización ambulatoria. El estudio también aborda los retos que plantean dispositivos implantables como marcapasos y desfibriladores; un estudio posterior propone una estrategia para eliminar pulsos distorsionados durante la reconstrucción, mejorando la calidad de la señal en cualquier condición. En conjunto, la tesis contribuye al avance de las metodologías de reconstrucción de derivaciones de ECG para mejorar la atención al paciente. / [CA] Aquesta tesi doctoral presenta un algoritme per a reconstruir el registre electrocardiogràfic (ECG) estàndard del sistema de 12 derivacions utilitzant un sistema reduït de derivacions independents mitjançant l'ús de models d'aprenentatge automàtic, centrant-se en la seua integració en un sistema de monitoratge ambulatori. Els mètodes tradicionals de reconstrucció de ECG es basen en enfocaments basats en combinacions lineals, amb una exploració limitada dels mètodes d'avaluació i de les posicions dels elèctrodes. Aquesta tesi avalua l'eficàcia de noves xarxes neuronals artificials i algoritmes basats en fuzzy c-means en comparació amb els mètodes clàssics de regressió lineal, destacant un rendiment superior i subratllant la importància de la explicabilitat del model. S'exploren altres millores, com a comités d'experts i models difusos, per a augmentar la precisió i l'eficàcia. La validació clínica realitzada a l'Hospital Clínic Universitari de València i a l'Hospital General Universitari de València demostren l'eficàcia de l'algoritme en la reconstrucció precisa de derivacions, facilitant el camí per a aplicacions de monitoratge ambulatori. L'estudi també aborda els reptes que plantegen dispositius implantables com a marcapassos i desfibril·ladors; un estudi posterior proposa una estratègia per a eliminar polsos distorsionats durant la reconstrucció, millorant la qualitat del senyal en qualsevol condició. En conjunt, la tesi contribueix a l'avanç de les metodologies de reconstrucció de derivacions de ECG per a millorar l'atenció al pacient. / [EN] This PhD Thesis presents an algorithm for reconstructing the standard 12-lead system electrocardiographic (ECG) register using a reduced system of independent leads supported by machine learning models, with a focus on its integration into an ambulatory monitoring system. Traditional ECG lead reconstruction methods have relied on linear combination based approaches, with limited exploration of evaluation methods and electrode positions. This thesis evaluates the effectiveness of new artificial neural networks and fuzzy c-means based algorithms compared to classical linear regression methods, highlighting superior performance and emphasizing the importance of model explainability. Further enhancements, including expert committees and fuzzy models, are explored to improve accuracy and efficiency. Clinical validation at the Hospital Clínico Universitario de València and Hospital General Universitario de València demonstrates the algorithm's effectiveness in an accurate lead reconstruction, paving the way for ambulatory monitoring applications. The study also addresses challenges posed by implantable devices such as pacemakers and defibrillators; a subsequent study proposes a strategy to eliminate distorted pulses during reconstruction, improving signal quality under any condition. Overall, the thesis contributes to advancing ECG lead reconstruction methodologies for improved patient care. / Grande Fidalgo, A. (2024). Design of an electrocardiographic lead reconstruction algorithm using machine learning in the context of ambulatory monitoring [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/214023
|
| 69 |
Ανάπτυξη τεχνικών επεξεργασίας ιατρικών δεδομένων και συστημάτων υποστήριξης της διάγνωσης στη γυναικολογίαΒλαχοκώστα, Αλεξάνδρα 25 May 2015 (has links)
Η αυτόματη επεξεργασία εικόνων του ενδομητρίου αποτελεί ένα δύσκολο και πολυδιάστατο πρόβλημα, το οποίο έχει απασχολήσει πλήθος ερευνητών και για το οποίο έχει αναπτυχθεί μεγάλος αριθμός τεχνικών.
Στην παρούσα διατριβή, παρουσιάζεται μια μεθοδολογική προσέγγιση, η οποία βασίζεται στη χρήση αλγορίθμων ψηφιακής επεξεργασίας και ανάλυσης εικόνων, για την αυτόματη εκτίμηση χαρακτηριστικών που περιγράφουν την αγγείωση και την υφή εικόνων του ενδομητρίου. Αφορμή της μελέτης αποτελεί ο ρόλος που διαπιστώνεται ότι διαδραματίζει η μεταβολή των τιμών των εν λόγω χαρακτηριστικών στην έγκαιρη διάγνωση των παθήσεων του ενδομητρίου.
Στα πλαίσια της διατριβής, υλοποιήθηκε κατάλληλη μεθοδολογία για τον υπολογισμό ενός συνόλου χαρακτηριστικών τόσο για υστεροσκοπικές εικόνες, όσο και για ιστολογικές εικόνες του ενδομητρίου. Ιδιαίτερη βαρύτητα δόθηκε στην προ – επεξεργασία των εικόνων προκειμένου να προκύψει βελτίωση της ποιότητας καθώς και ενίσχυση της αντίθεσης αυτών. Στη συνέχεια, ανιχνεύτηκαν τα σημεία που αποτελούν τους κεντρικούς άξονες των υπό εξέταση αγγείων με χρήση διαφορικού λογισμού για τις υστεροσκοπικές εικόνες και υπολογίστηκε ένα σύνολο χαρακτηριστικών μεγεθών που περιγράφουν την αγγείωση και την υφή των εικόνων τόσο για τις υστεροσκοπικές όσο και για τις ιστολογικές εικόνες. Τέλος, εφαρμόστηκαν κατάλληλοι αλγόριθμοι με σκοπό την κατηγοριοποίηση των υστεροσκοπικών και των ιστολογικών εικόνων και συγκεκριμένα τον διαχωρισμό των παθολογικών και των φυσιολογικών εικόνων του ενδομητρίου. Παράλληλα, χρησιμοποιήθηκε η ROC ανάλυση στην απεικόνιση και ανάλυση της συμπεριφοράς των εν λόγω κατηγοριοποιητών. / Automatic analysis of the endometrial images is a difficult and multidimensional problem. For this reason, the number of papers and techniques regarding this issue is numerous.
In this Thesis, a methodology is presented, based on advance image processing techniques in order to automatically estimate texture and vessel’s features in endometrial images. Motivation for the Thesis is the fact that the variation of the measurements of the specific features plays significant role in the seasonable diagnosis of endometrial disorders.
Throughout this Thesis, an appropriate methodology is developed in order to estimate the features for the hysteroscopical and histological images of the endometrium. An important step is the pre – processing of the images in order to enhance the image quality and the image contrast. Then, the pixels that constitute the centerlines of vessels are detected by using differential calculus for the hysteroscopical images, only. Furthermore, the texture and vessel’s features in hysteroscopical and histological images are estimated. Finally, appropriate algorithms are applied in order to classify the hysteroscopical and histological images and distinguish pathological and normal endometrial images. ROC analysis is used in order to evaluate the discrimination power of the features that were estimated.
|
| 70 |
Channel Probing for an Indoor Wireless Communications ChannelHunter, Brandon 13 March 2003 (has links) (PDF)
The statistics of the amplitude, time and angle of arrival of multipaths in an indoor environment are all necessary components of multipath models used to simulate the performance of spatial diversity in receive antenna configurations. The model presented by Saleh and Valenzuela, was added to by Spencer et. al., and included all three of these parameters for a 7 GHz channel. A system was built to measure these multipath parameters at 2.4 GHz for multiple locations in an indoor environment. Another system was built to measure the angle of transmission for a 6 GHz channel. The addition of this parameter allows spatial diversity at the transmitter along with the receiver to be simulated. The process of going from raw measurement data to discrete arrivals and then to clustered arrivals is analyzed. Many possible errors associated with discrete arrival processing are discussed along with possible solutions. Four clustering methods are compared and their relative strengths and weaknesses are pointed out. The effects that errors in the clustering process have on parameter estimation and model performance are also simulated.
|
Page generated in 0.0359 seconds