FCT: DF - Dissertações de Mestrado
URI permanente para esta coleção:
Navegar
Entradas recentes
- Simulation-based Medical Emergency training system for Deep-Space missions: A situation room approachPublication . Alves, Susana Filipa Dias; Lapão, LuísFuture deep-space missions, such as those in NASA’s Artemis program, will require astronauts and Mission Control to handle complex medical emergencies with limited support from Earth. This dissertation details CHIRON, a simulation-based training platform designed to prepare crews for high-stress medical scenarios in deep space missions. Built on a decision-support situation-room model, CHIRON incorporates relevant information, including real-time vital sign monitoring, scenario branching, and decision-making tasks, to assess both technical skills (medical and procedural knowledge) and non-technical skills (teamwork, situation awareness, communication, and task load). A pilot experiment was implemented during an analog lunar mission to test the ability of astronaut and Mission Control teams to manage simulated ischemic stroke during extravehicular activity in a Lunar Gateway mission. Results showed varied performance: astronauts excelled in procedural tasks, while Mission Control had stronger medical knowledge. Overall knowledge scores were below 70%, with survey delays, limited situational awareness, and inconsistent adherence to protocols. The NASA Task Load Index indicated a high workload, with astronauts feeling more frustrated and Mission Control citing time pressure. Teamwork was a strength, but weaknesses in leadership and task management persisted. These findings demonstrate CHIRON’s feasibility as both a research and training tool. By exposing critical gaps in clinical reasoning, workload management, and communication under operational stress, the platform provides a structured pathway to refine medical protocols and strengthen preparedness for autonomous care in space. Beyond astronaut training, CHIRON also holds promise for Earth-based disaster medicine, where teams must operate under resource scarcity, communication barriers, and high cognitive load.
- Optimização duma nova unidade de medição da força de contração do músculo Orbicularis OculiPublication . Guerreiro, Catarina Baptista; Ribeiro, Paulo; Duarte, AnaO músculo Orbicularis Oculi (OO) é responsável pelo encerramento das pálpebras e prote- ção do globo ocular. Existem várias patologias que podem afetar o seu funcionamento que tanto podem ser por um comprometimento direto do músculo (miopatias, traumatismos, cirurgias) ou pela sua enervação (paralisia do nervo facial idiopática ou secundária central ou periférica). É uma condição observada com relativa frequência na prática clínica. Nesta dissertação é proposto o desenvolvimento de uma plataforma de medição bio- mecânica do fecho palpebral, de forma a diagnosticar as diversas patologias associadas ao enfraquecimento do músculo OO ou mesmo a sua paralisia total. A unidade a desenvolver é composta por duas células de carga, cada uma associada a um módulo acondicionador de sinal, um microcomputador Raspberry Pi e um touchscreen para permitir a interação do utilizador (profissional de saúde) com uma interface user-friendly. Para além disso tem ainda um motor de passo e o seu controlador para permitir um posicionamento simples do dispositivo em relação ao paciente e uma bateria que alimenta tanto estas componentes anteriormente mencionadas como a restante eletrónica com o auxílio de um conversor DC-DC. Concluiu-se que esta unidade permite adquirir medições sistemáticas da força de contração do músculo OO para possibilitar a criação de uma base de dados e estudar as patologias anteriormente mencionadas. As várias medições da força de contração do músculo Orbicularis Oculi de um mesmo indivíduo são coerentes. Para além disso, tal como se suspeitava, cada indivíduo apresenta um padrão próprio. Contudo, os valores registados entre os vários voluntários estão na mesma gama de valores.
- Statistical Emulation of Complex Cardiac Models Using Gaussian ProcessesPublication . Amaro, Diogo Alexandre Inácio; Gao, Hao; Vigário, RicardoCardiovascular diseases remain the leading cause of mortality worldwide, and patient- specific cardiac computational models offer critical insights for diagnosis, prognosis, and the development of individualized treatment plans. The increasing sophistication of these models places them well beyond the reach of analytical solutions, demanding the use of numerical physics simulators to approximate physiological behavior. Such models are typically governed by complex partial differential equations (PDEs), whose solution requires computationally intensive numerical methods, making them impractical for real-time clinical applications. An alternative approach which can help bypass these challenges is emulation, whereby the original simulator is replaced by a data-driven surrogate model. This thesis explores the use of Gaussian Processes (GPs) as a statistical emulator for the passive diastolic filling of the left ventricle (LV), enabling fast approximation of model outputs from physiological input parameters. We illustrate the principles of surrogate modelling using two toy examples. The first highlights the underlying mechanics of GPs by fitting a sinusoidal function, while the second adapts this mechanism to a simplified cardiac scenario. Building on this foundation, we construct statistical surrogates for six output features generated by a high- fidelity forward simulation of left ventricular dynamics. These emulators are then used to perform predictive modelling and validated by Sobol sensitivity analysis. Lastly, we use those to perform parameter inference, both from a deterministic and probabilistic point of view. Finally, we interpret the results and demonstrate via theoretical analysis and numerical experiments that cardiac mechanical models can be deployed for real-time clinical decision support.
- Engineering Aligned Ice-Templated Scaffolds via Additive ManufacturingPublication . Fernandes, Ana Patrícia Rodrigues; Almeida, Henrique; Oliveira, JoãoBone tissue engineering offers promising alternatives to traditional bone grafting by de- veloping scaffolds that replicate the extracellular matrix and support tissue regeneration. This work explores how the thermal conductivity and surface geometry of metallic sub- strates, aluminum, copper, and stainless steel, used as cold surfaces in ice templating, influence the pore architecture and performance of gelatin-based scaffolds. A custom additive-manufactured device was developed to enable controlled direc- tional freezing and scaffold fabrication. Scaffolds were characterized morphologically, mechanically, and biologically using SEM, compression testing, porosity tests, swelling measurements, FTIR, and in vitro assays. Copper substrates, with the highest thermal conductivity, produced scaffolds with smaller, more aligned pores and higher compressive strength, while stainless steel and aluminum generated scaffolds with larger pores and higher porosity. All anisotropic scaf- folds displayed a microporous architecture, with an average pore size of approximately 16.8 μm across all groups, with pore orientation strongly dictated by substrate proper- ties. Although pore sizes were below the range typically reported as optimal for vas- cularization, the scaffolds showed high porosity, good swelling capacity, and mechanical properties within the range of cancellous bone. Cell viability assays confirmed short-term biocompatibility, with over 80% viability across all groups. These findings demonstrate that substrate material is a key parameter for modulating scaffold architecture in ice templating and provide mechanistic insights into how freezing conditions shape scaffold formation, contributing to the broader development of tailored biomaterials for bone tissue engineering research.
- The Acute Effects of Acoustic Biofeedback on Muscle PowerPublication . Guerreiro, Beatriz Vicente; Silva, Luís; Mira, JoséMuscle power training is often avoided in vulnerable populations, such as older adults, due to safety concerns associated with explosive movements. Yet, the age-related decline in muscle power, a condition known as powerpenia, represents the earliest and most critical loss of muscular function, increasing the risk of falls. To address the challenge of safely delivering power training to those who need it most, Acoustic Biofeedback (ABF) derived from muscle electrical activity can be leveraged to guide training that targets muscle power. This study aimed to investigate the acute effects of electromyography (EMG)-based ABF on muscle power in young adults, assessing the feasibility of the tool, as an ini- tial step toward its application in aging populations. Twenty-four healthy participants performed a power test with bench press and squat exercises at multiple relative loads across three sessions, under two auditory conditions: ABF and generic Gym Music (GM). Power-related and EMG-derived metrics were extracted, complemented by participants’ subjective enjoyment of the power test under both stimuli. Results showed that ABF elicited short-term improvements in men’s bench press performance, particularly in Maximum Power (𝐵𝐹−0 = 19.31) and Maximum Velocity (𝐵𝐹−0 = 9.198), whereas women displayed stable values largely influenced by physical activity level rather than stimulus type. Intramuscular coordination of upper and lower limbs did not change over sessions but revealed clear sex-related disparities. Although perceived enjoyment was higher with GM in both sexes, only ABF demonstrated acute performance advantages. These findings support EMG-based ABF as a promising aid for power-oriented training, suggesting that longer interventions could elicit further beneficial adaptations in muscular power output.
- New algorithms for pigmentation abnormalities detection in human skinPublication . Silva, Liliana Cristina dos Santos; Vieira, Pedro; Pinto, Pedro; Silva, PedroDetecting and localising regions of hyperpigmentation on facial skin is a challenging task in dermatology. These lesions are often diffuse and irregular, making them difficult to detect and track over time. Accurate localisation and classification of these regions are crucial for dermatological assessments and for companies like PhD Trials®, which evaluate the efficacy of topical products. The ability to extract precise detection metrics can significantly enhance the evaluation of treatment effectiveness, providing deeper insights into skin response. This dissertation presents a deep learning approach for developing a robust automatic object detection system for brownish hyperpigmentation on facial skin. Several pre- trained models, including YOLO and Faster R-CNN, were explored, along with different frameworks such as MATLAB and Detectron2. The proposed solution employs transfer learning to fine-tune these models with a custom dataset, optimising them for detecting and distinguishing lentigines from nevi. The research further extends the utility of these models by integrating segmentation using Otsu’s method, which enabled the delineation of the regions of interest and the extraction of monitoring metrics for each detected area, offering a more comprehensive evaluation of skin health. The results show that this solution outperforms PhD Trials®’ current software, providing superior detection and enhanced usability. The tool facilitates more reliable analyses in studies involving topical treatments, which are valuable for both clinical and cosmetic evaluations. Its ability to work with high-resolution camera images, not just dermoscopic ones, makes it suitable for broader clinical use, including telemedicine. This advancement represents a significant contribution to dermatological research and the development of AI-driven skin analysis tools.
- Mimicking animal behavior: Implementing reinforcement learning strategies on a wireless robotPublication . Pereira, Francisco Bastos; Marques, Hugo; Vieira, Pedro; McNamee, DanielAnimals constantly modify their behavior based on the outcomes of previous experiences; a process we commonly call learning. This work seeks to replicate certain biological learning processes in a mobile robotic platform, as the study of this processes in both biological and artificial systems has long been a subject of interest in neuroscience and robotics. In particular, the research focuses on operant conditioning, a learning mechanism in which an agent’s behavior is shaped by associating actions with a reinforcement, or punishment stimuli. In the mammalian brain, the basal ganglia is believed to play a key role in this process, implementing a kind of reinforcement learning framework that guides behavior through dopaminergic signals from the midbrain. To explore these principles, the study will utilize a wireless mobile robot developed at the Champalimaud Foundation. The robot will be trained to chase or avoid “objects” according to whether these have previously provided positive or negative feedback. These "objects" consist of images projected onto the ground by a ceiling-mounted projector and detected by the robot using a wireless camera. The ultimate objective is for the robot to learn associations between visual inputs and behavioral responses, ensuring it moves towards rewarding stimuli while avoiding negative ones. Throughout the project, we implement Reinforcement Learning techniques, specifically Temporal-Difference Learning. The programming of the robot as well as the experimental platform will is done by using Python and Bonsai, a reactive-based visual programming language, capable of naturally handling asynchronous streams of data.
- LOW TOTAL ELECTRON YIELD N-DOPED GRAPHENE COATINGS PRODUCED BY ELECTROPHORETIC DEPOSITIONPublication . Guerreiro, Pedro Miguel Alves; Bundaleski, NenadLow secondary electron emission (SEE) materials have gained significant attention in recent years, particularly in aerospace technologies, such as RF devices for satellite communications, and in fundamental particle physics studies. In high-intensity particle accelerators such as the High Luminosity Large Hadron Collider (HL-LHC) at CERN, materials with low SEE are crucial for mitigating electron cloud formation, thereby enabling higher maximum beam intensities, improved thermal conditions, and more reliable experimental results. This phenomenon results from the multipacting effect: electrons are generated either through ionization of the residual gas or via the photoelectric effect induced by synchrotron radiation. These high-energy electrons subsequently collide with the accelerator walls, producing secondary electrons. The electric field, produced by the bunches of positive particles, accelerates these secondary electrons, triggering a localized avalanche. The most common strategy to reduce the electron cloud formation is to coat the inner walls of accelerators with low Total Electron Yield (TEY) materials. Graphene is a promising candidate due to its inherently low SEE and surface roughness, which further reduces TEY. Recent studies revealed that nitrogen doping into graphene can lower its Secondary Electron Yield (SEY) even further, while also mitigating aging effects caused by repeated exposure to air. In this project, N-doped graphene free-standing powder was deposited onto metallic substrates, similar to those used as accelerator inner walls, using electrophoretic deposition (EPD). This technique was chosen for its ability to preserve the material’s original properties while offering a simple and cost-effective deposition method. Following the sample preparation, X-ray Photoelectron Spectroscopy (XPS) was used to characterize the samples in terms of surface chemical composition, while TEY measurements were performed. This allowed the identification of optimal deposition parameters and enabled the establishment of correlations between material characteristics and the TEY performance.
- ENCAPSULAMENTO DE UM INIBIDOR DE REPARAÇÃO DE ADN EM LIPOSSOMAS PARA TRATAMENTO DO CANCROPublication . Oliveira, Inês Alexandra dos Santos; Raposo, Maria de Fátima; Vieira, TâniaNeste trabalho pretendeu-se aprimorar o conhecimento de um tratamento inovador ba- seado na inibição de reparação do ácido desoxirribonucleico (ADN) através da utilização develiparib, uma molécula inibidora de uma proteína de reparação de ADN a Poli (ADP-ribose) Polimerase-1 (PARP1). Para tal, as moléculas de veliparib foram encapsuladas em lipossomas para que a entrega deste inibidor às células cancerígenas seja mais eficaz. Para se avaliar o efeito do inibidor e do inibidor encapsulado foram realizados estudos de citotoxicidade e fo- totoxicidade em duas linhagens de células. No desenvolvimento deste projeto as moléculas develiparib foram encapsuladas em lipossomas que foram preparadas a partir da mistura de 1,2- dipalmitoil-sn-glicero fosfocolina (DPPC) com 1,2 – dipalmitoil-sn-glicero-3-fosfoglicerol (DPPG). As soluções aquosas de veliparib e as emulsões de veliparib encapsulado em liposso- mas de diferentes concentrações, foram utilizadas em ensaios in vitro de citotoxicidade da molécula de veliparib em células cancerígenas denominadas de carcinoma de células escamo- sas (MET-1) e em células não cancerígenas de queratinócitos (HaCaT) e estudos de fototoxici- dade irradiadas com UV. Estes testes permitiram analisar a viabilidade celular após a exposição a diferentes concentrações do fármaco e o impacto da radiação UV na resposta celular. Os resultados indicam que o veliparib encapsulado em lipossomas é citotóxico nas células MET-1 e ao mesmo tempo reduz efeitos adversos nas células HaCaT sugerindo maior seletividade do tratamento. Estes resultados demonstram que esta abordagem pode reduzir a toxicidade nos tecidos saudáveis, melhorar a qualidade de vida dos pacientes e oferecer outra perspetiva de terapia anticancerígena.
- Evaluating the Potential of Gold-coated Nanodiamonds as Radiosensitizers to Enhance the Effectiveness of Radiation TherapyPublication . Alves, Gil Duarte Cabrita Dias; Belchior, Ana; Cruz, JoãoCancer remains one of the leading causes of death worldwide, demanding continuous advancements in therapeutic strategies. Radiotherapy is a widely used treatment modality, but its effectiveness is often limited by radioresistant cancer cells and collateral damage to healthy tissues. To address this issue, radiosensitizers have emerged as a promising solution to enhance the effectiveness of radiotherapy by locally increasing the radiation- induced damage to cancer cells, while minimizing harm to the surrounding healthy tissues. In this context, gold nanoparticles have demonstrated a considerable radiosensitizing potential, while nanodiamonds have been explored for their fluorescence properties. Accordingly, recent investigations suggest that by combining these materials into gold- coated nanodiamonds (NDAus), it may be possible to integrate radiosensitization with bioimaging, enabling theranostic applications. In this study, the radiosensitizing potential of NDAus was assessed in lung cancer cells exposed to 𝛾-rays, X-rays and protons, delivered by a cobalt-60 irradiator, a clinical photon beam and a Van de Graaff accelerator, respectively. The cell survival response was evaluated using clonogenic assays, while fluorescence microscopy was employed to investigate mechanisms of DNA damage and repair, reactive oxygen species production and lipid droplets formation. Results demonstrated a predominant radiosensitizing effect at an NDAus concentration of 10 𝜇g/mL under X-ray irradiation, with significant cell effects observed in the DNA damage and repair dynamics, as well as in the reactive oxygen species levels. This work was distinguished with the Young Investigator Award at the 49𝑡 ℎ conference of the European Radiation Research Society and was presented orally at both this meeting and the 11𝑡 ℎ Congress of Proteção Contra Radiações dos Países de Língua Portuguesa. Addi- tionally, the work was further discussed in the PIANOFORTE intensive course “Particle Irradiation: Molecular, Cellular and Tissue Effects”.