Orientador: Aparecido Antonio Camacho / Resumo: O objetivo do presente estudo foi investigar a função miocárdica pela técnica de ecocardiografia bidimensional speckle tracking (2D STE) em pacientes submetidos à procedimentos anestésico prolongado, com a intenção de se detectar possível disfunção sistólica. Inicialmente para se determinar o intervalo normal da função sistólica na espécie, foram utilizados 31 coelhos da raça Nova Zelândia, adultos, hígidos, machos, com peso médio de 4,0 ± 0,53 kg, nos quais foram obtidos os índices sistólicos radiais como, velocidade rotacional e radial, deslocamento rotacional e radial, strain radial e strain rate radial. Posteriormente, para investigação da função sistólica sob diferentes modalidades ventilatórias, 32 coelhos foram distribuídos, aleatoriamente, em quatro grupos de modalidades ventilatórias diferentes constituídos de 8 animais cada, denominados grupo ventilação ciclada à pressão (GP), grupo ventilação ciclada à pressão associada à PEEP (GPP), grupo ventilação mandatória intermitente sincronizada (GM) e grupo ventilação espontânea (GE). Em todos os grupos, os parâmetros foram mensurados 40 minutos após a indução anestésica (M0) e reavaliados a cada vinte minutos, durante duas horas ininterruptas (M1, M2, M3 e M4). Portanto, o resultado demonstrados na presente investigação cientifica revelam que nenhuma modalidade ventilatória estudada proporcionou um quadro de disfunção sistólica a luz da técnica 2D STE quando foram comparados em cada momento de avaliação anestésica. / Abstract: The aim of this thesis was to investigate the cardiac function of patients undergoing prolonged anesthesia with the use of a two-dimensional speckle tracking (2D STE) technique to detect systolic dysfunction in order to maintain safety during anesthetic procedure. The study was carried out in healthy rabbits because this species is widely used as a model of experimentation for both humans and other animal species. Initially, the normal range of left ventricular radial systolic function was measured, for this purpose, 31 healthy male adults New Zealand rabbits were used, with average weight of 4.0 ± 0.53 kg, from a specialized producer; radial systolic indices were obtained, such as rotational and radial speed, rotational and radial displacement, radial strain and strain rate. Subsequently, 32 rabbits were randomly assigned to four groups of different ventilatory modalities consisting of 8 animals each, referred to as pressure-cycling ventilation group (GP), pressure-cycling ventilatory group associated with PEEP (GPP), synchronized intermittent mandatory ventilation group (GM) and spontaneous ventilation group (GE). In all groups, the parameters were measured 40 minutes after anesthetic induction (M0) and reevaluated every twenty minutes, for two uninterrupted hours (M1, M2, M3 and M4). Therefore, the results demonstrated in the present scientific investigation reveal that no studied ventilatory modality provided a picture of systolic dysfunction in light of the 2D STE techni... (Complete abstract click electronic access below) / Doutor

Left ventricular diastolic dysfunction (LVDD) is any abnormality in the filling of the left ventricle (LV).  Despite the prevalence of this disease, it remains difficult to diagnose, mainly due to inherent compensatory mechanisms and a limited physical understanding of the filling process.  LV filling can be non-invasively imaged using color m-mode echocardiography which provides a spatio-temporal map of inflow velocity.  These filling patterns, or waves, are conventionally used to qualitatively assess the filling pattern, however, this work aims to physically quantify the filling waves to improve understanding of diastole and develop robust, reliable, and quantitative parameters. This work reveals that LV filling waves in a normal ventricle act as dispersive waves and not only propagate along the length of the LV but also spread and disperse in the direction of the apex.  In certain diseased ventricles, this dispersion is limited due to changes in LV geometry and wall motion.  This improved understanding could aid LVDD diagnostics not only for determining health and disease, but also for distinguishing between progressing disease states. This work also identifies a limitation in a current LVDD parameter, intra ventricular pressure difference (IVPD), and presents a new methodology to address this limitation.  This methodology is also capable of synthesizing velocity information from a series of heartbeats to generating one representative heartbeat, addressing inaccuracies due to beat-to-beat variations.  This single beat gives a comprehensive picture of that specific patient's filling pattern.  Together, these methods improve the clinical utility of IVPD, making it more robust and limiting the chance for a misdiagnosis. / Master of Science

The echocardiographic assessment of valvular heart diseases is the basic analysis of valvular defects next to clinical investigation and stethoscopy. Severity of mitral regurgitation (MR) is usually estimated by an integrated approach using semi quantitative parameters and is still one of the biggest challenges of echocardiography. Quantitative echocardiographic analysis of MR severity often fails to describe comprehensible hemodynamic conditions. However, comprehensive echocardiography based on standardized image acquisition and proper image quality is required to properly assess hemodynamic parameter comparable to cardiac magnetic resonance tomography. This review focuses on the uncertainty of MR severity assessed by echocardiography in recent trials of interventional MR treatment. In addition, the necessity to provide plausible echocardiographic data for individual decision making is highlighted. In conclusion, plausible functional diagnostics by rational echocardiography is a prerequisite in patients with valvular heart diseases.

Simulation based education has been shown to increase the task-specific capability of medical trainees. Transesophageal echocardiography training greatly benefits from the use of simulators. They allow real time scanning of a beating heart and generation of ultrasound images side by side with anatomically accurate virtual model. These simulators are costly and have many limitations. 3D printing technologies have enabled the creation of bespoke phantoms capable of being used as task-trainers. This study aims to compare the ease of use and accuracy of a low-cost patient-specific, Computer-tomography based, 3D printed, echogenic TEE phantom compared to a commercially available echocardiography training mannequin. We hypothesized that a low-cost, 3D printed custom-made, cardiac phantom has comparable image quality, accuracy and usability as existing commercially available echocardiographic phantoms. After Institutional Ethic Research Board approval, we recruited ten American Board – Certified cardiac anesthesiologists and conducted a blinded comparative study divided into two stages. Stage one consisted of image assessment. A set of basic TEE views obtained from the 3D printed and commercial phantom were presented to the participants on a computer screen in random order. For each image, participants will be asked to identify the view, identify the quality of the image on a 1-5 Likert scale compared to the corresponding human view and guess with which phantom it was acquired (1 not at all realistic to patients view and 5 realistic to patients view). Stage two, participants will be asked to use the 3D printed and the commercially available phantom to obtain basic TEE views. In a maximum of 30 minutes. Each view was recorded and assessed for accuracy by two certified echocardiographers. Time needed to acquire each basic view and number of correct views was recorded. Overall usability of the phantoms was assessed through a questionnaire. For all continuous variables, we will calculate mean, median and standard deviation. We use Wilcoxon Signed-Rank test to assess significant differences in the rating of each phantom. All ten participants completed all part of the study. All participants could recognize all of the standard views. The average Likert scale was 3.2 for the 3D printed and 2.9 for the commercial Phantom with no significant difference. The average time to obtain views was 24.5 and 30 sec for the 3D printed and the commercial phantoms respectively statistically significantly in favor of the 3D printed phantom. The qualitative user assessment for ease to obtain the views, probe manipulation, image quality and overall experience were in great favor of the 3D printed phantom. Our Study suggest that the quality of TEE images obtained on the 3D printed phantom are not significantly different from those obtained on the commercial Phantom. The ease of use and time required to complete a basic TEE exam were in favor of the 3D Printed phantom.:Table of Content 1. Bibliographic Description 3 2. Introduction 4 2.1. Perioperative transesophageal echocardiography 4 2.2. Transesophageal echocardiography training 5 2.3. Transesophageal echocardiography simulation 6 2.4. 3D Heart Printing 13 2.5. 3D Segmentation 16 2.6. Development of the study phantom 17 2.7. Study Rationale 18 3. Publication 22 4. Summary 30 5. References 33 6. Appendices 37 6.1. Darstellung des eigenes Beitrags 38 6.2. Erklärung über die eigenständige Abfassung der Arbeit 39 6.3. Lebenslauf 40 6.4. Publikationen und Vorträge 44 6.5. Danksagung 61

Background Medically managed tricuspid regurgitation (TR) has detrimental outcomes. Transcatheter tricuspid valve interventions (TTVIs) represent an alternative to surgery in high-risk patients; however, only early experiences exist. Aim The aim of this study was to analyze the clinical and echocardiographic outcomes of TTVI. Methods MEDLINE, ISI Web of Science, and SCOPUS databases were searched for studies published up to June 2021. Studies reporting data on outcome post-TTVIs were included. This study was designed according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) requirements. The primary endpoint was all-cause mortality at 30-day and 1-year post-TTVI. Results Out of 2,718 studies, 27 were included. Notably, 30-day and 1-year all-cause mortalities were 5% (95% confidence interval [CI]: 4–8%, p < 0.001) and 25% (95% CI: 12–45%, p = 0.016). Procedural success was associated with a 58% risk reduction in 1-year mortality vs. lack thereof (odds ratio 0.42, 95% CI: 0.27–0.66, p < 0.001). TTVI is associated with a significant reduction in TR severity (TR EROA, mean difference [MD] 0.31 cm2; 95% CI: 0.23–0.39 cm2, p < 0.001; regurgitant volume, MD 23.54 ml; 95% CI: 17.4–29.68 ml, p = 0.03) and increase in forward stroke volume (FSV, MD 3.98 ml; 95% CI: 0.11–7.86 ml, p = 0.04). Conclusion TTVI significantly reduces TR severity and increases FSV and is associated with improved survival at 1 year compared with patients without procedural success. Long-term outcomes compared with medical therapy await the results of ongoing pivotal trials; nonetheless, TTVIs appear to be a promising alternative to surgery for TR.

No description available.

No description available.

No description available.

No description available.

Данная работа исследует методы обучения U-Net для сегментации эхокардиографических изображений и показывает возможный вариант внедрения модели. Работа основана на анализе эффективности и применимости машинного обучения в сегментации эхокардиографических изображений. Исследование описывает возможные методы машинного обучения, а также метрики для задачи сегментации. В работе рассматривается разработка веб-приложения для автоматизации сегментирования эхокардиографических изображений в медицинских центрах, которое позволит ускорить процесс приема пациентов, не потеряв качества сегментации. / This work explores U-Net training methods for segmentation of echocardiographic images and shows a possible implementation of the model. The work is based on the analysis of the effectiveness and applicability of machine learning in segmentation of echocardiographic images. The study describes possible machine learning methods, as well as metrics for the segmentation task. The paper considers the development of a web application for automating segmentation of echocardiographic images in medical centers, which will speed up the process of receiving patients without losing the quality of segmentation.