<p dir="ltr">This dissertation addresses three progressively fundamental problems for functional data analysis: (1) To do efficient inference for the functional mean model accounting for within-subject correlation, we propose the refined and bias-corrected empirical likelihood method. (2) To identify functional subjects potentially from different populations, we propose the dominant-set based unsupervised clustering method using the similarity matrix. (3) To learn the similarity matrix from various similarity metrics for functional data clustering, we propose the modularity guided and dominant-set based semi-supervised clustering method.</p><p dir="ltr">In the first problem, the empirical likelihood method is utilized to do inference for the mean function of functional data by constructing the refined and bias-corrected estimating equation. The proposed estimating equation not only improves efficiency but also enables practically feasible empirical likelihood inference by properly incorporating within-subject correlation, which has not been achieved by previous studies.</p><p dir="ltr">In the second problem, the dominant-set based unsupervised clustering method is proposed to maximize the within-cluster similarity and applied to functional data with a flexible choice of similarity measures between curves. The proposed unsupervised clustering method is a hierarchical bipartition procedure under the penalized optimization framework with the tuning parameter selected by maximizing the clustering criterion called modularity of the resulting two clusters, which is inspired by the concept of dominant set in graph theory and solved by replicator dynamics in game theory. The advantage offered by this approach is not only robust to imbalanced sizes of groups but also to outliers, which overcomes the limitation of many existing clustering methods.</p><p dir="ltr">In the third problem, the metric-based semi-supervised clustering method is proposed with similarity metric learned by modularity maximization and followed by the above proposed dominant-set based clustering procedure. Under semi-supervised setting where some clustering memberships are known, the goal is to determine the best linear combination of candidate similarity metrics as the final metric to enhance the clustering performance. Besides the global metric-based algorithm, another algorithm is also proposed to learn individual metrics for each cluster, which permits overlapping membership for the clustering. This is innovatively different from many existing methods. This method is superiorly applicable to functional data with various similarity metrics between functional curves, while also exhibiting robustness to imbalanced sizes of groups, which are intrinsic to the dominant-set based clustering approach.</p><p dir="ltr">In all three problems, the advantages of the proposed methods are demonstrated through extensive empirical investigations using simulations as well as real data applications.</p>

Le mélanome est la forme la plus grave de cancer de la peau. Cette thèse a contribué au développement de deux approches différentes pour le diagnostic assisté par ordinateur du mélanome : approche analytique et approche comparative.L'approche analytique imite le comportement du dermatologue en détectant les caractéristiques de malignité sur la base de méthodes analytiques populaires dans une première étape, et en combinant ces caractéristiques dans une deuxième étape. Nous avons étudié l’impacte d’un système du diagnostic automatique utilisant des images dermoscopique de lésions cutanées pigmentées sur le diagnostic de dermatologues. L'approche comparative, appelé concept du Vilain Petit Canard (VPC), suppose que les naevus chez le même patient ont tendance à partager certaines caractéristiques morphologiques ainsi que les dermatologues identifient quelques groupes de similarité. VPC est le naevus qui ne rentre dans aucune de ces groupes, susceptibles d'être mélanome. / Melanoma is the most serious type of skin cancer. This thesis focused on the development of two different approaches for computer-aided diagnosis of melanoma: analytical approach and comparative approach. The analytical approach mimics the dermatologist’s behavior by first detecting malignancy features based on popular analytical methods, and in a second step, by combining these features. We investigated to what extent the melanoma diagnosis can be impacted by an automatic system using dermoscopic images of pigmented skin lesions. The comparative approach, called Ugly Duckling (UD) concept, assumes that nevi in the same patient tend to share some morphological features so that dermatologists identify a few similarity clusters. UD is the nevus that does not fit into any of those clusters, likely to be suspicious. The goal was to model the ability of dermatologists to build consistent clusters of pigmented skin lesions in patients.

[ES] La fibrilación auricular (FA) es la arritmia cardíaca más común y afecta a más de 33 millones de pacientes en el mundo. A menudo se encuentra en unidades de cuidados intensivos, donde se asocia con hospitalizaciones prolongadas, mayores costos de atención médica, riesgo elevado de tromboembolismo y mayor mortalidad. La FA tiene diversas causas y mecanismos y se considera una enfermedad heterogénea. Puede ser causada por comorbilidades cardíacas y no cardíacas, como trastornos endocrinos, pulmonares y metabólicos, genética e inflamación. La abundancia de mecanismos fisiopatológicos asociados con la FA ha llevado a la comprensión de que los pacientes con FA son considerablemente heterogéneos. Esta heterogeneidad entre las poblaciones de pacientes se ha identificado previamente como un impedimento no abordado en los estudios epidemiológicos. Existen pautas para el tratamiento y manejo de la FA para la población general, pero no son directamente aplicables a las poblaciones de la UCI debido a los diferentes mecanismos, riesgos y efectividad de los tratamientos de la FA. Además, falta evidencia sólida sobre estrategias de tratamiento óptimas, lo que resulta en una falta de consenso entre los tomadores de decisiones clínicas y diferentes enfoques de tratamiento en las instituciones clínicas. Esta tesis doctoral informa el proceso de desarrollo de un método de estratificación para pacientes con FA en el entorno de cuidados críticos. Se desarrollan, comparan y emplean nuevos algoritmos de agrupamiento semisupervisados para identificar fenotipos de FA. Se comparan los efectos del tratamiento de fármacos antiarrítmicos comunes entre fenotipos y se realiza una evaluación de usabilidad para identificar la aplicabilidad clínica de los métodos desarrollados. / [CA] La fibril·lació auricular (FA) és l'arítmia cardíaca més comú i afecta més de 33 milions de pacients al món. Sovint es troba en unitats de cures intensives, on s'associa amb hospitalitzacions prolongades, majors costos d'atenció mèdica, risc elevat de tromboembolisme i més mortalitat. La FA té diverses causes i mecanismes i es considera una malaltia heterogènia. Pot ser causada per comorbiditats cardíaques i no cardíaques, com ara trastorns endocrins, pulmonars i metabòlics, genètica i inflamació. L'abundància de mecanismes fisiopatològics associats a la FA ha portat a la comprensió que els pacients amb FA són considerablement heterogenis. Aquesta heterogeneïtat entre les poblacions de pacients s'ha identificat prèviament com un impediment no abordat als estudis epidemiològics. Hi ha pautes per al tractament i maneig de la FA per a la població general, però no són directament aplicables a les poblacions de la UCI a causa dels diferents mecanismes, riscos i efectivitat dels tractaments de la FA. A més, manca evidència sòlida sobre estratègies de tractament òptimes, la qual cosa resulta en una manca de consens entre els prenedors de decisions clíniques i diferents enfocaments de tractament a les institucions clíniques. Aquesta tesi doctoral informa el procés de desenvolupament d'un mètode d'estratificació per a pacients amb FA a l'entorn de cures crítiques. Es desenvolupen, comparen i fan servir nous algorismes d'agrupament semisupervisats per identificar fenotips de FA. Es comparen els efectes del tractament de fàrmacs antiarítmics comuns entre fenotips i es fa una avaluació d'usabilitat per identificar l'aplicabilitat clínica dels mètodes desenvolupats. / [EN] Atrial fibrillation (AF) is the most commonly encountered cardiac arrhythmia, affecting over 33 million patients in the world. It is often encountered in intensive care units, where it is associated with prolonged hospitalisation, increased healthcare costs, elevated risk of thromboembolism, and higher mortality. AF has diverse causes and mechanisms, and is considered to be a heterogeneous disease. It may be caused by cardiac and non-cardiac comorbidities, such as endocrine, pulmonary, and metabolic disorders, genetics, and inflammation. The abundance of pathophysiological mechanisms associated with AF has led to the realization that AF patients are considerably heterogeneous. This heterogeneity among patient populations have previously been identified as an unaddressed impediment in epidemiological studies. Guidelines for the treatment and management of AF exist for the general population but are not directly applicable to ICU populations due to different AF mechanisms, risks, and effectiveness of treatments. Further, strong evidence for optimal treatment strategies is missing, resulting in a lack of consensus among clinical decision-makers, and different treatment approaches across clinical institutions. This doctoral thesis reports the process of developing a stratification method for AF patients in the critical care setting. Novel semi-supervised clustering algorithms are developed, benchmarked, and employed to identify AF phenotypes. Treatment effects of common antiarrhythmic drugs are compared among phenotypes, and a usability assessment is performed to identify the clinical applicability of the developed methods. / Lacki, AS. (2024). Clinical Decision Support System for the Multiparametric Stratification of Atrial Fibrillation Patients in Critical Care [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/212511