Atrial arrhythmias are the most prevalent sustained cardiac arrhythmias. Rates of hospitalisation and costs incurred to healthcare organisations are increasing in epidemic proportions. Despite this, the mechanisms of the transition from sinus rhythm to arrhythmic states are not well understood. The high level of regional electrical heterogeneity observed in the atria is thought to contribute towards the high prevalence of atrial arrhythmias. However, current computer models of the intact human atria only account for a small degree of this regional electrical heterogeneity, and do not include descriptions of the pacemaker regions of the sinoatrial node and the atrioventricular node. In this project, a new computational model of the intact 3D human atria is developed. First, a new single cell model to simulate the electrical action potential of the human atrial myocyte is developed. This model more accurately simulated the experimentally observed properties of human atrial action potentials than previous models. A family of electrically heterogeneous models describing the major regions within the atria is then developed, including those of the sinoatrial- and atrioventricular- nodes. This set of regional cell models represents the most expansive and complete set currently available. It is demonstrated that the large range of different electrical properties results in a large range of action potential morphology and duration within the atria. Models of the effect of sympathetic and parasympathetic regulation on the electrical AP of the models of the atrial working myocardium and the pacemaker regions were also incorporated. This demonstrated that sympathetic regulation can increase the pacing rate of the sinoatrial node and the atrio-ventricular node, and has a complex dose dependent effect on the atrial working myocardium. Four distinct models of the effects of atrial fibrillation induced remodelling on the atrial working myocardium are developed. These characterised the effect of remodelling of IKur on the overall changes in action potential morphology and duration observed. It is shown that the presence or absence of remodelling of this channel accounts for two distinct observed morphologies. A previous 3D anatomical model of the human atria is improved. First, detailed anatomical models for the sinoatrial node and the atrioventricular node are incorporated into the model. Second, it is further segmented to include regions for the pulmonary veins, atrio-ventricular ring, atrial septum and sinoatrial node block zone. This model is used to investigate the effects of sympathetic and parasympathetic regulation in the 3D atria. Finally, a detailed investigation of the underlying mechanisms of atrial fibrillation in the 3D atria, and the effect of electrical remodelling on the behaviour of atrial fibrillation, is performed using the detailed 3D model. This work represents a significant advance in 3D human atrial modelling. The anatomical model incorporates a greater level of complexity than previous models, and for the first time allowed investigation of the pacemaking mechanisms in the 3D intact human atria. The atrial fibrillation protocols are more physiologically relevant than previous models and have elucidated the roles that electrophysiological remodelling, electrical heterogeneity and structural anisotropy play in the development and maintenance of atrial fibrillation.

Introduction: Atrial fibrillation is the most common sustained arrhythmia. The quality of atrial fibrillation treatment in clinical practice varies between hospitals and doctors. Due to the increasing prevalence of atrial fibrillation in the more elderly population, it might be difficult in the future to refer patients to a limited number of heart specialists. At the same time, a new generation of University educated nurses is growing, but, unfortunately, their competencies are as low as their colleagues' with high-school education. However, in some areas, such as diabetology or gastroenterology, specialist nurses, whose aim is to supplement medical doctors to some extent, are gradually occurring. In the Czech Republic, there are no specialist nurses in cardiology, even though they are quite common abroad. Our aim is to assess the feasibility of a specialist nurse for diagnosis and treatment of atrial fibrillation. Methods: Main inclusion criterion for our study was a diagnosis of atrial fibrillation made on one of the non-internal medicine wards or in the general practitioners' practice. An exclusion criterion was previous examination by one of the heart specialists either internal medicine doctor or a cardiologist. During the first interview, I checked previous investigations for the atrial fibrillation and the arrhythmia treatment. The questionnaire contained questions on demographic parameters and patients' knowledge about their disease and possible complications. Subsequently, I recorded my recommendations regarding further investigations and therapeutic possibilites. All patients were sent to a cardiologist as required by our national guidelines. The second interview was made over the phone 2 months later and comparison of the doctor's recommendations with my suggestions was made. During this interview, an assessment of the concept of a nurse specialist in atrial fibrillation was performed. Results: Forty patients with atrial fibrillation documented on an ECG were recruited. At a time of the first interview, only 10 patients (25%) had their echocardiogram performed and only 23 (57,5%) patients had their thyroid gland hormones checked. 26 (65%) patients were on anticoagulation medicine while CHADS2-Vasc score of 2 and more was documented in 34 (85%) patients. There was an agreement between doctor's and nurse's recommendations regarding missing investigations and anticoagulation treatment. However, no agreement was achieved in the area of arrhythmia treatment, which means that this should remain a doctor's responsibility. Patients were informed about possible complications in only 11 (27,5%) cases and about all therapeutic options in 2 (5%) cases. 29 (72,5%) patients would be interested in consulting a specialist nurse in the future, whereas 11 (27,5%) patients would prefer a doctor. Results: We managed to confirm that the novel concept of a specialist nurse for the diagnosis and treatment of atrial fibrillation is feasible. The specialist nurse is able to confirm the diagnosis, check the completeness of the investigations, educate patients and assess the risk of the thromboembolic disease. Competencies of the nurses should be extended, so this concept could be introduced into the clinical practice.

An integration of ICT advances into a conventional healthcare system is spreading extensively nowadays. This trend is known as Electronic health or E-Health. E-Health solutions help to achieve the sustainability goal of increasing the expected lifetime while improving the quality of life by providing a constant healthcare monitoring. Cardiovascular diseases are one of the main killers yearly causing approximately 17.7 million deaths worldwide. The focus of this work is on studying the detection of one of the cardiovascular diseases – Atrial Fibrillation (AF) arrhythmia.  This type of arrhythmia has a severe influence on the heart health conditions and could cause congestive heart failure (CHF), stroke, and even increase the risk of death. Therefore, it is important to detect AF as early as possible. In this thesis we focused on studying various machine learning techniques for AF detection using only short single lead Electrocardiography recordings. A web-based solution was built as a final prototype, which first simulates the reception of a recorded signal, conducts the preprocessing, makes a prediction of the AF presence, and visualizes the result. For the AF detection the relatively high accuracy score was achieved comparable to the one of the state-of-the-art. The work was based on the investigation of the proposed architectures and the usage of the database of signals from the 2017 PhysioNet/CinC Challenge. However, an additional constraint was introduced to the original problem formulation, since the idea of a future deployment on the resource-limited devices places the restrictions on the complexity of the computations being performed for achieving the prediction. Therefore, this constraint was considered during the development phase of the project.

The topic of my diploma thesis is: ?The Questions of the Care of Patients before and after the Intervention for Ventricular Fibrillation ?. The diploma thesis defines three objectives. The first objective finds out the information level of patients about nursing care before and after the intervention for ventricular fibrillation. The second objective was to find out the most frequent problems in connection with the intervention for ventricular fibrillation. The third objective aims at surveying the requirements for the nursing care in connection with the intervention for ventricular fibrillation. To conduct my research I defined 4 questions along with these objectives: Are patients informed about nursing care before and after the intervention for ventricular fibrillation? In which way do nurses inform patients about nursing care before and after the intervention for ventricular fibrillation? How do nurses assess the cooperation with patients before and after the intervention for ventricular fibrillation? In which fields are patients most limited after the intervention for ventricular fibrillation? To fulfil the objectives of the diploma thesis I chose the quantitative research survey. The collection of data was conducted by the interview technique. These interviews constituted the basis for the chart of ideas in the X-mind programme. The research questions that were defined for the research survey yielded these conclusions: The information on the regime before the radiofrequency ablation for fibrillation of chambers is insufficient for the patients. The information on the regime after the radiofrequency ablation for ventricular fibrillation is insufficient for the patients. Nurses inform patients about the nursing care before the performance based on nursing standards. The largest problem encountered by patients after the radiofrequency ablation for ventricular fibrillation is bowel movement. Patients before the radiofrequency ablation for ventricular fibrillation worry because of lack of information about the performance. The patients before the radiofrequency ablation for ventricular fibrillation cooperate with the nursing staffs. The patients after the radiofrequency ablation for ventricular fibrillation cooperate with the nursing staffs.

Introdução: O tromboembolismo é uma grave complicação da fibrilação atrial (FA), particularmente em pacientes que vão se submeter à cardioversão, química ou elétrica. Para reduzir esse risco, os pacientes submetem-se à anticoagulação clássica, que vem sendo praticada há várias décadas. Apesar desta abordagem, em pacientes plenamente anticoagulados, não se conhece a prevalência de trombo ou contraste espontâneo no átrio esquerdo (AE). Por essa razão, alguns autores sugerem a realização do ecotransesofágico (ECOTEE) para confirmar o sucesso do tratamento e reduzir o risco de complicações tromboembólicas após a reversão. Os objetivos deste estudo foram: a) avaliar a prevalência de trombos e contraste espontâneo ao ECOTEE em pacientes que vão ser submetidos à cardioversão sob regime de anticoagulação plena; b) avaliar a incidência de tromboembolismo até 30 dias após o procedimento; c) avaliar a influência das variáveis clínicas (doenças associadas) e do ECOTEE (tamanho e volume indexado do AE, fração de ejeção ventricular; velocidade de fluxo no apêndice atrial esquerdo), além do escore CHA2DS2VASc e níveis de pró-BNP plasmático sobre a formação de trombo/contraste espontâneo. Métodos: Foram incluídos 85 pacientes (62 homens; média de idade 61±12 anos) com FA não valvar com indicação para cardioversão. Todos receberam varfarina com controle da taxa de INR. Quando se considerava o paciente plenamente anticoagulado (INR ente 2 e 3 por três semanas consecutivas), era prescrito um fármaco antiarrítmico (propafenona, sotalol ou amiodarona) cuja escolha se baseou em critérios clínicos. Na ausência de normalização do ritmo, eram encaminhados para cardioversão elétrica (CVE). No dia da CVE, os pacientes submetiam-se ao ECOTEE cujo resultado só era conhecido no dia seguinte após a cardioversão. Os pacientes recebiam alta com anticoagulante e retornavam ao ambulatório após 30 dias quando realizavam outro ECOTEE. Resultados: Todos os pacientes foram cardiovertidos com INR na faixa terapêutica (2,9±0,7). A reversão com fármacos ocorreu em 9/85 pacientes (10,6%); 67/76 pacientes submeteram-se à CVE e, destes, 58/67 (86%) reverteram ao ritmo sinusal. O ECOTEE antes da CVE evidenciou trombo no AE em 8/85 pacientes (9,4%) e contraste espontâneo em 36/85 pacientes (42,3%). Nenhuma variável clínica, escore CHA2DS2VASc, níveis plasmáticos de pró-BNP ou variáveis ecocardiográficas identificou pacientes com maior probabilidade de apresentar trombo/contraste espontâneo no AE. Após 30 dias, houve normalização das variáveis do ECOTEE. Em 5/8 (62,5%) pacientes, os trombos desapareceram e surgiu em outros dois pacientes (2,3%). O contraste espontâneo desapareceu em 24/38 (63%) pacientes. Não houve registro de nenhum caso de tromboembolismo sistêmico em 30 dias. A taxa de recorrência de FA foi de 21%. Conclusões: a) trombo atrial/contraste espontâneo foi detectado em 9,4% da população e nenhuma variável clínica ou ecocardiográfica identificou pacientes de risco; b) houve melhora das variáveis do ECOTEE após a reversão ao ritmo sinusal; d) o sucesso global da cardioversão foi de 88% e a taxa de recorrência de FA de 21% em 30 dias; c) não houve registro de tromboembolismo sistêmico em 30 dias, em ritmo sinusal ou em FA. / Introduction: Thromboembolism is a serious complication of atrial fibrillation (AF), particularly in patients who will undergo chemical or electrical cardioversion. To reduce this risk patients receive classic anticoagulant therapy, which has been practiced for several decades. Despite this approach, it is not known the prevalence of thrombus or spontaneous contrast in the left atrium (LA) in patients fully anticoagulated. For this reason, some authors have recommended the transesophageal echocardiogram (TEECHO) to reduce the risk of thromboembolic complications after cardioversion. The objectives of this study were: a) to evaluate the prevalence of thrombus and spontaneous contrast by TEECHO in patients about to undergo cardioversion under full anticoagulation regime; b) evaluate the incidence of thromboembolism within 30 days after the procedure; c) evaluate the influence of clinical variables (associated diseases) and TEECHO parameters (LA size and LA indexed volume, ventricular ejection fraction, flow velocity in the left atrial appendage), CHA2DS2VASc score and plasma pro-BNP levels on thrombus/spontaneous contrast formation. Methods: We included 85 patients (62 men; mean age 61 ± 12 years) with non-valvular AF referred for cardioversion. All received warfarin with INR control. When considering the patient fully anticoagulated (INR in the range of 2 to 3 for three weeks) it was prescribed an anti-arrhythmic drug (propafenone, sotalol or amiodarone) whose choice was based on clinical criteria. In the absence of normal rhythm, patients were referred for electrical cardioversion (ECV). On the day of ECV, all patients were submitted to the ECOTEE whose result was known only the next day after cardioversion. The patients were discharged with anticoagulant and returned to the clinic after 30 days when another ECOTEE was performed. Results: All patients were cardioverted with INR in the therapeutic range (2.9±0.7). Sinus rhythm was restored with drugs in 9/85 patients (10.6%); 67/76 patients underwent ECV and 58/67 (86%) reverted to sinus rhythm. The TEECHO before cardioversion showed a thrombus in LA in 8/85 patients (9.4%) and spontaneous contrast in 36/85 patients (42.3%). No clinical variable, CHA2DS2VASc score, pro-BNP plasma levels or echocardiography variables identified patients with an increased likelihood of thrombus/spontaneous contrast in LA. After 30 days, there was normalization of TEECHO variables. In 5/8 (62.5%) patients thrombi disappeared and appeared in two patients (2.3%). Spontaneous contrast disappeared in 24/38 (63%) patients. There were no reports of any case of systemic thromboembolism in 30 days. The AF recurrence rate was 21%. Conclusions: a) LA thrombus/ spontaneous contrast were detected in 9.4% of the population and no clinical or echocardiography variable identified patients at risk; b) there was an improvement of TEECHO variables after reversion to sinus rhythm; d) the overall success of cardioversion was 88% and the AF recurrence rate was 21% in 30 days; c) there was no systemic thromboembolism in 30 days, in patients in sinus rhythm or AF.

No description available.

This thesis proposes the use of Machine Learning (ML) to two very distinct, yet compelling, applications – predicting cardiac arrhythmia episodes and predicting base station association in 5G networks comprising of virtual cells. In the first scenario, Support Vector Machines (SVMs) are used to classify features extracted from electrocardiogram (EKG) signals. The second problem requires a different formulation departing from traditional ML classification where the objective is to partition feature space into constituent class regions. Instead, the intention here is to identify temporal patterns in unequal-length sequences. Using Recurrent Neural Networks (RNNs), it is demonstrated that accurate predictions can be made as to the base station most likely to provide connectivity for a mobile device as it moves. Atrial Fibrillation (AF) is a common cardiac arrhythmia affecting several million people in the United States. It is a condition in which the upper chambers of the heart are unable to contract effectively leading to inhibited blood flow to the ventricles. The stagnation of blood is one of the major risk factors for stroke. The Computers in Cardiology Challenge 2001 was organized to further research into the prediction of episodes of AF. This research revisits the problem with some modifications. Patient-specific classifiers are developed for AF prediction using a different dataset and employing shorter EKG signal epochs. SVM classification yielded an average accuracy of just above 95% in identifying EKG epochs appearing just prior to fibrillatory rhythms. 5G cellular networks were envisaged to provide enhanced data rates for mobile broadband, support low-latency communication, and enable the Internet of Things (IoT). Handovers contribute to latency as mobile devices are switched between base stations due to movements. Given that customers may not be willing to continuously share their exact locations due to privacy concerns and the establishment of a mobile network architecture with dynamically created virtual cells, this research presents a solution for proactive mobility management using RNNs. A RNN is trained to identify patterns in variable-length sequences of Received Signal Strength (RSS) values, where a mobile device is permitted to connect to more than a single base station at a time. A classification accuracy of over 98% was achieved in a simulation model that was set up emulating an urban environment.

Abstract Atrial fibrillation (AF) is a common arrhythmia in the clinical setting with a population prevalence of 1–2%. AF significantly increases the risk of stroke and death, worsens coexistent heart diseases and may leave the patient with disabling symptoms. The treatment of AF consists of the control of the underlying conditions, prevention of complications and symptom relief by controlling heart rate (rate control) or by targeting normal rhythm (rhythm control), with the latter achieved either by antiarrhythmic drug (AAD) therapy or catheter ablation (CA). Ablation therapy has generally been applied and studied after failure of AAD therapy. The aim of this study was to evaluate the safety and efficacy of first-line CA in AF. The other objectives were to assess the safety of continuous warfarin therapy during CA and to identify prognostic markers for treatment outcome. A meta-analysis of all randomised studies and a secondary analysis of one randomised study comparing CA and AAD as first-line therapy were performed. In the first study, ablation therapy reduced AF recurrences more than AAD therapy (HR 0.63) when provided as first-line therapy; the rate of complications was similar with both therapies. Some of the complications of ablation therapy were more serious than those encountered with AADs. The second study revealed that the antiarrhythmic efficacy of ablation therapy was more durable. In the third study, the efficacies of continuous and interrupted warfarin therapy were compared in 228 procedures; both strategies were found to be equally safe during a three month follow-up. Furthermore, an analysis of 2317 AF episodes revealed a new electrocardiographic feature at AF initiation, which was associated with AF relapse after the initiation of therapy. Finally, a thorough echocardiographic examination was performed in 49 patients prior to ablation therapy. Mild diastolic dysfunction was associated with AF recurrence. In conclusion, CA was more effective as a first-line therapy than AADs but may cause more severe complications. Continuous warfarin therapy was found to be safe during CA. New electrocardiographic and echocardiographic markers for treatment outcome were recognised. / Tiivistelmä Eteisvärinä on yleinen rytmihäiriö, jonka esiintyvyys väestössä on 1–2 % luokkaa. Eteisvärinä lisää merkittävästi kuolleisuutta ja aivoinfarktiriskiä, vaikeuttaa muiden sydänsairauksien oireita ja saattaa aiheuttaa invalidisoivia oireita. Eteisvärinän hoito keskittyy liitännäissairauksien hoitoon ja komplikaatioiden estoon sekä oireiden lievitykseen joko syketaajuutta säätämällä (sykkeenhallinta) tai pyrkimällä normaaliin rytmiin (rytminhallinta). Rytminhallinnassa käytetään yleisesti joko rytmihäiriölääkkeitä tai katetriablaatiohoitoa. Eteisvärinän katetriablaatiota on useimmiten käytetty ja tutkittu tilanteessa, jossa rytmihäiriölääkitys on osoittautunut tehottomaksi. Tämän tutkimuksen tavoitteena oli arvioida eteisvärinän katetriablaatiohoidon tehoa ja turvallisuutta ensilinjan hoitona. Muina tavoitteina oli katetriablaation turvallisuuden arviointi jatkuvan varfariinihoidon aikana sekä löytää uusia katetriablaatiohoidon tehoa ennustavia tekijöitä. Teimme meta-analyysin kaikista randomisoiduista tutkimuksista ja sekundaarisen analyysin yhdestä randomisoidusta tutkimuksesta, jotka vertasivat rytmihäiriölääke- ja katetriablaatiohoitoa ensilinjan hoitona. Ensimmäisessä työssä ablaatiohoito esti eteisvärinän uusiutumista tehokkaammin (riskisuhde 0.63), eikä komplikaatioiden yleisyydessä ollut eroa hoitojen välillä. Jotkut ablaatiohoitoon liittyvät komplikaatiot olivat kuitenkin luonteeltaan vakavampia kuin lääkehoidossa. Ablaatiohoidon eteisvärinää estävä vaikutus todettiin pidempikestoiseksi toisessa työssämme. Kolmannessa työssä vertasimme jatkuvaa ja tauotettua varfariinihoitoa 228 ablaatiotoimenpiteen aikana. Molemmat lähestymistavat osoittautuivat yhtä turvallisiksi 3 kuukauden seuranta-aikana. Analysoimme edelleen 2317 eteisvärinäkohtausta ja löysimme osalla potilaista uuden eteisvärinäkohtauksen alkuun liittyvän ominaisuuden, joka oli yhteydessä rytminhallinnan tehottomuuteen. Lisäksi teimme 49 potilaalle laajan sydämen ultraäänitutkimuksen ennen katetriablaatiotoimenpidettä. Diastolisen dysfunktion havaittiin olevan yhteydessä eteisvärinän uusiutumiseen. Yhteenvetona totesimme että katetriablaatiohoito on rytmihäiriölääkehoitoa tehokkaampaa ensilinjan hoitona, mutta siihen mahdollisesti liittyvät komplikaatiot olivat luonteeltaan hankalampia. Jatkuva varfariinihoito todettiin turvalliseksi katetriablaation yhteydessä. Löysimme lisäksi sydänsähkökäyrästä ja sydämen ultraäänitutkimuksesta uusia hoidon tehoa ennustavia tekijöitä.

Background: Atrial fibrillation (AF) is the most prevalent cardiac arrhythmia. There is a need for new pharmacological treatment strategies since the current antiarrhythmic drugs have a modest efficacy and may have severe side effects. Cardioversion (CV) of AF offers an opportunity to study related conditions in sinus rhythm (SR) and during AF. Since catheter ablation of AF is a symptomatic treatment, it is important to have tools for measurement of arrhythmia-related symptoms. Aims: To evaluate the effect of atorvastatin on maintaining SR after CV of persistent AF. To assess if highsensitivity C-reactive protein (hsCRP) predicts the recurrence of AF after CV in a population randomized to treatment with either atorvastatin or placebo. To quantify the symptomatic effect of left atrial catheter ablation of AF. To assess if the restoration of SR by CV, in a population with persistent AF, affects sleep apnea. Methods: Paper I: A total of 234 patients were randomized to treatment with either high dose atorvastatin or placebo prior to CV. Paper II: In a pre-specified substudy which included 128 of the patients in study I, hsCRP was analyzed before and after CV. Paper III: Umea 22 Arrhythmia Questions (U22) is a questionnaire that quantifies paroxysmal tachycardia symptoms. A total of 105 patients underwent first-time pulmonary vein isolation and answered U22 forms at baseline and follow-up 304 (SD 121) days after ablation. Paper IV: Polysomnography was performed before and after CV in 23 patients with persistent AF scheduled for elective CV. Results: Paper I: An intention-to-treat analysis with the available data, by randomization group, showed that 57 (51%) in the atorvastatin group and 47 (42%) in the placebo group were in SR 30 days after CV (OR 1.44, 95%CI 0.85–2.44, P=0.18). Paper II: HsCRP did not significantly predict recurrence of AF at 30 days. However, after adjusting for treatment with atorvastatin, hsCRP predicted the recurrence of AF (OR 1.14, 95% CI 1.01–1.27). Six months after CV, hsCRP at randomization predicted recurrence of AF in both univariate analysis (OR 1.30, 95% CI 1.06–1.60) and in multivariate logistic regression analysis (OR 1.33, 95% CI 1.06– 1.67). Paper III: The U22 scores for well-being, arrhythmia as cause for impaired well-being, derived timeaspect score for arrhythmia, and discomfort during attack detected relevant improvements of symptoms after the ablation. U22 showed larger improvement in patients undergoing only one procedure than in patients who later underwent repeated interventions. Paper IV: Obstructive sleep apnea occurred in 17/23 patients (74%), and central sleep apnea in 6/23 patients (26%). Five patients had both obstructive and central sleep apnea. SR at follow-up was achieved in 16 patients. The obstructive apnea-hypopnea index, central apneahypopnea index, and the number of patients with obstructive or central sleep apnea did not differ before and after restoration of SR. Conclusions: Atorvastatin is not a treatment option with regards to maintaining SR after CV in patients with persistent AF. HsCRP was associated with AF recurrence 1 and 6 months after successful CV of persistent AF. U22 quantifies the symptomatic improvement after AF ablation with adequate internal consistency and construct validity. Both obstructive and central sleep apneas are highly prevalent in patients with persistent AF. Obstructive sleep apneas are unaffected by the CV of AF to SR.

Responsable d’un quart des accidents vasculaires cérébraux, la fibrillation auriculaire (FA) est l’arythmie cardiaque la plus répandue. La thérapie d’ablation par cathéter (CA) est de plus en plus utilisée pour traiter la FA, mais ses effets sur le substrat cardiaque ne sont pas suffisamment compris, d’où un taux de réussite très variable. L’électrocardiogramme (ECG) à 12 voies représente un outil non invasif peu coûteux pour caractériser la FA à partir de l’activité électrique du cœur. Cependant, les prédicteurs classiques de l’issue de la CA présentent plusieurs inconvénients, notamment leur calcul manuel sur une seule voie de l’ECG. Cette thèse exploite explicitement le caractère multi-capteur de l’ECG au moyen de techniques de décomposition multivariées, démontrant qu’elles peuvent améliorer la puissance prédictive de certaines propriétés de l’ECG dans le cadre de la CA. L’amplitude des ondes fibrillatoires est corrélée avec le résultat de la CA, et traitée par une méthode multi-capteur basée sur l’analyse en composantes principales (PCA). Des variantes comme la PCA pondérée (WPCA) et la factorisation en matrices non négatives (NMF) peuvent aussi quantifier la variabilité spatio-temporelle de la FA sur l’ECG. La théorie de l’information permet également d’estimer le niveau de corrélation entre les voies de l’ECG, mis en relation avec le résultat de la CA grâce à des approches multi-capteurs. Enfin, une dernière ligne de recherche concerne la réponse ventriculaire manifestée sur la variabilité cardiaque. L’approche paramétrique de processus ponctuel est capable de souligner certaines propriétés de cette variabilité, améliorant ainsi la caractérisation de la FA. / Atrial fibrillation (AF) is the most common sustained arrhythmia encountered in clinical practice, and one of the main causes of stroke. Yet its thorough characterization and treatment remain an open issue. Despite the increasing popularity of the radiofrequency catheter ablation (CA) therapy, very little is known about its impact on heart substrate, leading to rather uncertain success rates. This calls for advanced signal processing tools for quantitatively assessing CA outcome. The surface 12-lead electrocardiogram (ECG), a noninvasive and cost-effective cardiac activity recording modality, provides valuable information about AF. However, some issues affect most of the standard CA outcome predictors, e.g., manual computation and limited single-lead perspective. This thesis aims at explicitly exploiting the ECG’s multilead character through multivariate decomposition tools, so as to enhance the role of some ECG features as CA outcome predictors. Fibrillatory wave amplitude is correlated with CA success in a multilead framework through principal component analysis (PCA). Multivariate approaches also enhance AF spatiotemporal variability measured on the ECG (e.g., weighted PCA, nonnegative matrix factorization), evidencing that the less repetitive the AF pattern, the less likely CA success. Information theory also quantifies interlead similarity between AF patterns, and is linked with CA outcome in a multilead framework. Another perspective focuses on the ventricular response as reflected on heart rate variability (HRV). Point process modeling can highlight certain HRV properties typical of AF in a parametric probabilistic context, helping AF pattern recognition.