Treballs Finals de Grau (TFG) - Enginyeria Biomèdica

URI permanent per a aquesta col·leccióhttps://diposit.ub.edu/handle/2445/147724

Treballs Finals del Grau de Dret d'Enginyeria Biomèdica de la Facultat de Medicina i Ciències de la Salut de la Universitat de Barcelona.

Estadístiques

Examinar

Mostrant 1 - 20 de 113

User-centered integrated redesign of the Electrophysiology Room at the Sant Joan de Déu Pediatric Hospital
(2025-06-11) Torres Expósito, Clàudia; Jordi Colomer i Farrarons
This project presents a comprehensive redesign proposal for the Electrophysiology Room at the Arrhythmia Unit of Sant Joan de Déu Children’s Hospital (HSJD). The objective is to resolve existing structural and functional inefficiencies through a needs-driven, paediatric-adapted design. The approach follows the Biodesign methodology, progressing from the identification of needs to the proposal of solutions. The first phase involved shadowing procedures, interviews with clinicians and nurses, detailed questionnaires and benchmarking visits to six Spanish hospitals. This led to the collection and classification of over 40 needs across four categories: infrastructure, equipment, patient experience, and staff comfort. Each need was prioritised (must-have, nice to have, or not needed) based on CSUR standards, clinical workflows, and national recommendations.These needs were translated into a set of finalessential and recommended technical requirements. For each, a tailored solution was proposed, integrating real-world observations and commercially available technologies. The proposal includes ceiling-mounted storage and power systems, improved lighting and temperature control, and better ergonomic conditions. All solutions were compiled into a full list of recommended equipment and illustrated in a conceptual layout plan.The project concludes that while technical design is key, subjectivity plays a major role in shaping what each team considers "ideal." Hence, the "perfect" EP room is not a universal standard but a dynamic solution that must respond to the unique needs of a clinical team at a specific point in time.
Automatic Feature Extraction Pipeline for Abdominal Aortic Aneurysm Characterization
(2025-06-11) Vila Delgado, Júlia; xxxx
The completion of this project would not have been possible without the help and support I received throughout its development. First, I would like to give my sincere thanks to Ager Uribezubia for his ongoing guidance and support throughout the entire project. I would also like to thank Dr. Josep Munuera for giving me the opportunity to be part of this project and for trusting my work. I am also grateful to César Acebes and the Dimension Lab team, whose help and advice allowed me to solve several challenges along the way. Finally, I want to thank my family and friends for always being there for me, giving me support, encouragement, and motivation throughout this journey. Thank you all for making this project possible
“MRI-based radiomics machine learning model for tumour response prediction to neoadjuvant chemoradiotherapy (nCRT) in locally advanced rectal cancer (LARC): a retrospective study
(2025-06-11) Sanahuja Rosich, Paula; Núria Gavara i Casas
Rectal cancer (RC) is one of the most commonly diagnosed malignant tumours worldwide. Its most aggressive form, Locally Advanced Rectal Cancer (LARC), is treated with neoadjuvant chemoradiotherapy (nCRT) to downstage the tumour and improve results of the total mesorectal excision (TME). Approximately 20% to 25% achieve complete response and may be eligible for a more conservative watch and wait strategy. Early prediction of response to nCRT using information from the staging MRI could help adjust neoadjuvant treatment andimprove response, potentially avoiding surgeries. The aim of this project is to develop a machine learning (ML) model capable of predicting response to nCRT using clinical data and radiomics characteristics extracted from the pre-treatment MRI. The radiomics are extracted from manually delineated tumour masks (Core Tumour Radiomics) and from border masks computed from the manual segmentations (Border Radiomics). Nine ML models have been optimized and tested across the seven feature set combinations. The models with best performance were Random Forest for the core tumour radiomics dataset (Accuracy = 0.77, AUC = 0.70, Sensitivity = 0.67, Specificity = 0.86) and Multilayer Perceptron for the dataset with all features (Accuracy = 0.77, AUC = 0.70, Sensitivity = 0.83, Specificity = 0.71). However, the predictive capability of tumour borders could not be confirmed as these models yielded worse performance. Furthermore, via a feature importance analysis, it has been concluded that both shape and texture related radiomic features are predictors of treatment response although no specific marker has been identified.
Development of a customizable website to follow-up physical rehabilitation of cardiovascular patients at home
(2025-06-10) Sabala de Gregorio, Ariadna; Ramon Farré Ventura
The growing demand for remote patient monitoring solutions, especially in the management of chronic conditions such as cardiovascular diseases, has highlighted the need for flexible and accessible digital tools for home monitoring. This project presents the development of an open source exercise monitoring website to support cardiac rehabilitation programs, specifically tailored for the East Galway Roscommon Integrated Care Hub in Ireland. The platform enables patients to complete a periodic diary of physical activities, including exercise type, duration, and intensity, among others, and allows them to view their progress through graphs. It also allows healthcare professionals to monitor the patients’ exercise data, send feedback messages, and export the data in a conventional spreadsheet format. The website was tested with simulated patients and professionals, with more than 2.000 data and messages transferred without error. Feedback confirmed the platform’s reliability, ease of use and compatibility with all major browsers. A key objective of the project was to make the tool implementable and adaptable by healthcare professionals without the need for web development skills. To achieve this, a comprehensive implementation and customization manual has been developed, allowing healthcare professionals in a variety of clinical settings to adapt the platform to their specific needs. Given that existing telemedicine tools are often generic and inflexible, this project presents a practical and low-cost solution for healthcare professionals to monitor the progress of their patients and offer them personalized support.
Design and Validation of a Mechanical Ventilator for Pandemic Emergencies in Low-Resource Settings Using Commonly Available Components
(2025-06-11) Ros Alonso, Joana; Ramon Farré Ventura
This project introduces the design and validation of a cost-effective mechanical ventilator explicitly developed for emergency use in low-resource settings during pandemics and compassive use. It aims to provide a practical, easy-to-manufacture solution in environments where traditional medical equipment is either unavailable or unaffordable due to logistical, economic, or supply chain barriers. The ventilator is primarily constructed from widely available automotive parts, including windshield wiper motors and bellows. Two versions of the device were created: one for adults and a smaller, adapted model for neonates. Core ventilation parameters (tidal volume, breathing rate, and airway pressure) can be adjusted using simple mechanical and electrical controls. Performance was evaluated through a series of tests using artificial lung models under various simulated clinical conditions. The results showed consistent delivery of target volumes and pressures, with reliable operation across a range of breathing frequencies. Although the ventilator is not certified for hospital use, it demonstrates significant potential as an emergency respiratory support tool in humanitarian crises. Its design emphasizes simplicity, affordability, and adaptability, offering a feasible alternative for resource-limited healthcare environments
Energy Optimization in Research Infrastructure: A Data-Driven Analysis of the CEK Building
(2024-06-11) Passada Gil, Júlia; Albert Compte
This thesis presents a bottom-up energy audit of the Esther Koplowitz Centre (CEK) building at IDIBAPS, Barcelona, to guide targeted energy-saving actions. Due to the absence of permanent metering, this study combined a detailed equipment inventory, short-term monitoring campaigns, and statistical modeling of hourly electricity data from 2023–2024. The calibrated model explains 97% of measured weekly demand, with a relative error of 3%, and captures seasonal variation with a Mean Absolute Percentage Error (MAPE) of 6.8%. Disaggregation reveals a concentrated energy profile, with HVAC systems responsible for ~52% of annual use, followed by laboratory equipment (~36%) and the Data Processing Center (CPD; ~9%). Regression analysis further shows that outdoor temperature and daily occupancy explain 83% of day-to-day energy variability, with summer temperatures strongly influencing seasonal peaks. Three high-impact interventions emerge, ranked by estimated savings: (i) submetering and recommissioning HVAC subsystems; (ii) raising set-points of ultra-low temperature (ULT) freezers from -80 °C to -70 °C; and (iii) increasing the CPD cooling set-point from 24 °C to 26 °C. Together, these measures would cut consumption by ≈ 0.43 GWh per year (about 8.1 MWh per week)—11 % of today’s 3.87 GWh annual load. Despite limited metering infrastructure, this approach demonstrates how a datainformed audit can reliably uncover savings opportunities and provide a scalable audit framework applicable to comparable biomedical research infrastructures.
Development of a Temperature- Controlled Environment for Neurovalidation
(2025-06-11) Mora Menéndez, Guillermo; Guillermo Mora
This Final Degree Project presents the design and implementation of a temperature-controlled chamber for studying the behavior of neural precursors under varying thermal conditions. The chamber is intended as a core component within a broader initiative aimed at developing non- invasive therapeutic devices for epilepsy treatment, particularly through ultrasound-induced brain cooling.The device enables precise environmental temperature control. Its design integrates heating elements, high-precision digital temperature sensors (TSIC 506F), and a closed-loop control system managed by an Arduino Uno microcontroller, as well as a LabVIEW-based interface developed for real-time data visualization and logging. A methacrylate structure, supported by 3D-printed components, was selected for its transparency and ease of fabrication after multiple versions were investigated to optimize chamber structure, thermal distribution, and sensor integration. Both passive and forced-convection heating methods were analyzed, with thermal validation conducted using infrared imaging. The chamber consistently achieved the desired thermal range (36–37.5 °C), demonstrating spatial uniformity in controlled experiments and representing an initial platform for analyzing the impact of temperature on neural activity, contributing valuable data for the design of future brain-cooling neuromodulation devices.
Transforming Hospital Clínic de Barcelona Logistics through RFID: An In-Depth Study of Implementation, Validation and Economic Assessment
(2025-06-11) Iglesias Yepes, Judit; Neus Vidal
fficient supply chain management plays a critical role in intrahospital logistics. Radiofrequency identification (RFID) technology has emerged as a powerful tool to optimize the supply chain across multiple industries, including healthcare. Its application in hospitals is rapidly expanding due to its ability to automate inventory tracking, reduce human error, and improve data accuracy. This final degree project is framed within the context of the RFID implementation project at Hospital Clínic de Barcelona (HCB), where the technology has been deployed in surgical and interventional areas. The system includes RFID tagging of high-value consumables, prostheses, and medical devices; with fixed RFID readers installed throughout operating rooms, storage areas, and waste disposal zones, and corridors. Implementing RFID in a complex hospital environment like HCB presents considerable logistical and technical challenges. As part of this thesis, I actively contributed to the implementation process. One of the main contributions was the successful validation of the shielding paint in adjacent rooms, which effectively eliminated signal interference. Moreover, a material compatibility study identified specific packaging types that interfered with radiofrequency signal propagation, allowing for corrective measures. Although the system is still in the final phase of implementation, early results are promising. The integrated RFID solution has demonstrated robust performance, significantly reducing non-value time spent by clinical staff on logistical tasks. An economic analysis conducted in one of the pilot sites showed a favourable return on investment (ROI), further supporting the system’s viability and potential scalability across the hospital.
Preprocessing and decoding neural traces of serial biases in working memory from intracranial electrophysiological recordings in humans
(2025-06-11) Hurtado Morell, Alberto; Albert Compte, Alexis Pérez
This thesis would not have been possible without the support of many people. I am deeply grateful to Albert Compte for giving me the opportunity to be a part of this research and for his mentorship and encouragement, and to Daniel Pacheco, whose help in iEEG cleaning has brought us to where we are now. Additionally, I wanted to express my gratitude to the Hospital Clinic and its Unit of Epilepsy, who allowed us to work closely with them and collect data. Most of all, I am deeply thankful to the patients who collaborated and made this research possible. I am also especially grateful to Dr. Roldán, whose assistance in providing patient information (MRI sequences, CT scans, and surgical planning details) was fundamental to this study. Furthermore, I would like to thank my professors, whose guidance and knowledge have shaped my academic journey, and my family and friends, who have been my source of support and encouragement throughout this process. Their patience, belief in me, and motivation have been of much help in completing this thesis. Finally, I would like to express my deepest gratitude to Alexis Pérez, whose mentorship, guidance, and support have been invaluable throughout this research. His patience, expertise, and willingness to help at every step of the analysis have made a significant impact on this work. I am truly grateful for his dedication and for everything I have learned from him during this journey.
Comparative analysis of intraoperative radiotherapy technologies
(2025-06-10) Hamdan BenKssouba, Yassir; Dr. Alberto Biete
Intraoperative radiotherapy (IORT) is a specialized technique where radiation is delivered directly to a tumour bed immediately after surgical removal, significantly reducing treatment duration and limiting damage to healthy tissue. Today, hospitals considering IORT face a challenging decision among electron-based systems (IOeRT), low-voltage X-ray units, high-dose-rate (HDR) brachytherapy, or conventional external-beam IMRT. Currently, there is no practical comparative guideline tailored to help hospital administrators select the most suitable system based on their specific clinical, logistical, and financial contexts. This thesis provides a comprehensive comparative analysis of these technologies by synthesizing information from an extensive review of the scientific literature, direct observations from leading hospitals in Barcelona, and structured interviews with medical physicists, radiation oncologists, surgeons, and healthcare managers. Each technology’s clinical indications, technical complexity, infrastructure demands, staffing needs, radiation safety requirements, and both initial and ongoing costs were systematically assessed. Findings reveal that IOeRT systems offer broad clinical versatility and short treatment times, though they require substantial infrastructure adjustments. Low-voltage systems offer simpler setup with minimal shielding but are restricted to specific, smaller surgical cavities. HDR brachytherapy excels in dose precision for irregular anatomical sites but adds operational complexity and ongoing source costs. IMRT provides extensive clinical applications but demands significant infrastructural investment and prolonged treatment schedules. Ultimately, this thesis serves as a practical resource, equipping hospital decision-makers with clear criteria to identify and implement the optimal IORT technology that aligns best with their institutional capabilities and patient care goals.
Development of a Clinical-Radiomic Model for Head and Neck Cancer Outcome Prediction Based on Multi-Scanner PET Imaging
(2025-06-11) González Cuesta, Nuria; Gabriel Reynés Llompart, Laura Rodríguez Bel
First of all, I would like to thank Aida Niñerola for her mentorship and guidance throughout this project. I am especially grateful for her constant willingness to help, clarify doubts, and manage the progress of the project. I would also like to thank Gabriel Reynés and Laura Rodríguez for their constant support, patience, and commitment. Their dedication has been essential for the development of this project. I also appreciate their generosity in sharing their knowledge and their advice, which has helped me grow professionally. I feel very grateful to have had the opportunity to work with talented professionals and to be part of a team defined by mutual care and passion for this field. I extend my thanks to all members of the Nuclear Medicine Department at Hospital Universitari de Bellvitge, who warmly welcomed me into their group from the very beginning, providing me with their support and help. Finally, I am very grateful to my family for their love and support throughout this project. Above all, I want to thank them for instilling in me the importance of culture, critical thinking, and perseverance.
AI-based diagnostic algorithm for cellular aging using single-cell fluorescence imaging of the cytoskeleton
(2025-06-11) Colomé Xicoy, Laia; Núria Gavara i Casas
L’envelliment és un procés complex i multifactorial caracteritzat per un declivi fisiològic, un augment del risc de malalties i un deteriorament funcional progressiu. Els fibroblasts, essencials per mantenir l'estructura i la reparació dels teixits, mostren canvis relacionats amb l’edat àmpliament documentats. Tot i això, identificar marcadors robustos i quantificables d’aquests canvis continua sent un repte. Aquest projecte desenvolupa un protocol per detectar l’envelliment cel·lular mitjançant imatges de fluorescència de cèl·lules individuals. Els fibroblasts de ratolins joves i vells són immunotenyits per a marcadors nuclears i citoesquelètics, i se n’obtenen imatges mitjançant microscòpia d’epifluorescència multicanal. Es desenvolupa un mètode computacional per segmentar les cèl·lules i extreure característiques que descriuen la morfologia, la intensitat, la textura i l’organització del citoesquelet. Les característiques que superen un primer filtratge estadístic s’utilitzen després per entrenar i validar models d’aprenentatge automàtic interpretables. Es selecciona la regressió logística per la seva combinació d’interpretabilitat i rendiment, assolint una precisió del 0.97 en el conjunt d’avaluació. El protocol final ofereix un mètode reproductible per detectar l’envelliment dels fibroblasts només a partir d’imatges de microscòpia. Permet tant el diagnòstic a partir d’una classificació binària com la comprensió mecanística dels canvis cel·lulars relacionats amb l’edat, i estableix les bases per a aplicacions futures com els rellotges d’envelliment, el descobriment de biomarcadors i l’avaluació de teràpies orientades a modular aquests canvis.
Use of Machine Learning and SNOMED CT Encoded Health Problems to Predict Hospital Discharge Diagnoses
(2025-06-11) Chen, Cindy; J. C. Barrios Montenegro
The accurate classification of discharge diagnoses is a critical step in clinical decision-making, as it has direct effect on patient care, hospital management, and administrative tasks. Traditionally, diagnostic coding has been a manual and time-consuming process, typically done after a patient is discharged, which could lead to delays for subsequent processes such as billing, reporting, and care optimization. Recently, the Hospital Clínic de Barcelona has integrated a structured list of health problems coded in SNOMED CT into the Electronic Health Record (EHR) from the beginning of the patient’s hospitalization. This development has enabled the reuse of structured clinical data throughout the care process and has opened the door for predictive tools using Machine Learning (ML). The goal of this research is to determine whether there’s a significant relationship between reported health problems and the final ICD-10 discharge diagnoses. To explore this, data obtained from the Hospital Clínic de Barcelona was analysed, incorporating information from various clinical sources, such as demographics, laboratory results, prescriptions, and admissions records. Feature engineering was also carried out and methods based on decision trees, along with ANOVA tests, were used to identify the most relevant input variables. Subsequently, several supervised ML models, including Decision Trees (DTs), Random Forest (RF), and XGBoost were trained and evaluated. The best performing model, a Decision Tree classifier, achieved an accuracy of 69.8%, with a recall and F1-score of 0.68, and an AUC of 0.83. While no single variable served as a dominant predictor, the results show that health problems coded in SNOMED CT, combined with other clinical and
Flexible fMRI Processing Pipeline for Large- Scale Studies of Functional Connectivity in Preclinical Alzheimer’s Disease
(2025-01-20) Biosca Marron, Marc; Roser Sala Llonch , Raúl Tudela Fernández
El cervell es pot entendre com una xarxa complexa de regions espacialment distribuïdes, però funcional- ment interconnectades que comparteixen informació contínuament. La connectivitat funcional (FC), que mesura com les regions del cervell es comuniquen i coordinen, pot ajudar a identificar signes inicials de malalties com l’Alzheimer (AD). Aquest projecte té com a objectiu implementar un pipeline flexible per processar dades de ressonància magnètica funcional en repòs (rs-fMRI) i estudiar la FC en etapes preclíniques de l’AD. Un conjunt de dades de més de 1.000 subjectes ha estat processat mitjançant un pipeline flexible, que consisteix en una sèrie de passos com l’extracció cerebral, la correcció de moviment i la reducció de soroll, dissenyats per preparar les dades per a posteriors anàlisis. A més, s’ha desenvolupat una aplicació de control de qualitat (QC), que combina la inspecció visual i mètriques quantitatives per garantir resultats fiables, amb 76% dels subjectes superant el QC. Tots els procediments desenvolupats en aquest projecte estan disponibles en un repositori de GitHub. Els resultats demostren una relació significativa entre la FC i l’escala Centiloid, una mesura de la progressió de l’AD. També s’han identificat canvis en la connectivitat cerebral relacionats amb l’edat, mostrant potenciació i deteriorament de connexions, indicant una redistribució de les xarxes cerebrals. Finalment, s’ha introduït un marc exploratori per analitzar altres factors que poden influir en la funció cerebral, obrint la porta perquè es pugui utilitzar en futurs estudis de neuroimatge. Investigacions fu- tures haurien d’avaluar el potencial d’aquests biomarcadors com a indicadors primerencs de malalties neurodegeneratives
Computer-aided design of orthoses for children with cerebral palsy A proof-of-concept study
(2025-06-11) Ávalo Gadea, Berta; Gutiérrez Gálvez, Agustín
Els infants amb paràlisi cerebral (PC) sovint presenten patrons de marxa alterats a causa de dèficits neuromusculars i habitualment se’ls prescriuen ortesis de turmell-peu (AFOs) com a part de la seva rehabilitació. Aquesta tesi contribueix al desenvolupament d’eines computacionals per donar suport a la presa de decisions clíniques, explorant l’ús de simulacions musculoesquelètiques predictives per al disseny personalitzat d’ortesis. El projecte investiga com diferents valors de rigidesa de les AFOs i diferents velocitats de marxa afecten la biomecànica del caminar en infants amb PC. La metodologia comença amb la personalització de models musculoesquelètics per a dos pacients pediàtrics amb PC, incorporant afectacions motores com la debilitat muscular i les contractures articulars. A continuació, s’afegeixen AFOs amb diversos nivells de rigidesa als models, i es realitzen simulacions predictives de manera iterativa per a diferents velocitats de marxa. Per interpretar els resultats de les simulacions, s’extreuen sis paràmetres biomecànics de cada simulació: rang de moviment, angle del turmell, torque total del turmell, activació muscular, cost de transport i longitud de la passa. Aquests indicadors s’utilitzen per avaluar la correlació entre el rendiment de la marxa específic del pacient i el rendiment normatiu d’un pacient sà, proporcionant una base per seleccionar les configuracions òptimes de l’ortesi. Finalment, es proposa un criteri de selecció per determinar la rigidesa més eficaç de l’AFO per a cada extremitat i la velocitat de marxa més adequada per a cada pacient. Aquest criteri s’aplica als dos casos d’estudi per recomanar prescripcions ortopèdiques individualitzades. Tot i limitacions com la mida reduïda de la mostra i l’elevada càrrega computacional, l’estudi dona suport a la viabilitat i rellevància clínica del disseny d’ortesis específiques per a cada pacient basat en simulacions.
Study of the effectiveness of semi-immersive and immersive Virtual Reality in rehabilitation patients
(2025-01-20) Auger García, Inés; Gerard Trias
Biomedical Engineering Inés Auger García 1 Abstract Virtual reality (VR) refers to the use of computer-generated environments that simulate real or imagined experiences. These environments are designed to be interactive, allowing users to engage with them in real-time. Based on the level of immersion, there are three kinds of virtual reality: non-immersive, semi-immersive and fully immersive. This study investigates the effectiveness of semi-immersive and immersive VR modalities in rehabilitation therapy, with a specific focus on their impact on patient’s pain evolution, recovery, and engagement. The primary objective is to assess the effectiveness of these modalities through a comprehensive analysis of patient data. Metrics such as pain perception, functional recovery, and patient feedback were evaluated to determine therapeutic outcomes. Additionally, other primary objectives like studying the integration of the VR in rehabilitation will also be done. This study was conducted in the Rehabilitation Department of Platon Hospital, Hospital Clinic in Barcelona. All phases involved, from the patient selection to the discussion of results, are detailed in this project. Keywords: Virtual Reality, Immersive VR, Semi-Immersive VR, VR Rehabilitation Therapy, Pain Management, Functional Recovery, VR Integration in Healthcare, Clinical Study. Resum La realitat virtual (RV) fa referència a l’ús d’entorns generats per ordinador que simulen experiències reals o imaginàries. Aquests entorns permeten als usuaris interactuar amb ells en temps real. Segons el grau d’immersió, hi ha tres modalitats diferents de realitat virtual: no immersiva, semi-immersiva i immersiva. L’objectiu principal d’aquest estudi és observar l’efectivitat de les modalitats de RV semi-immersiva i immersiva en la teràpia de rehabilitació. Durant l’estudi s’avaluaran diferents mètriques com l’evolució del dolor i la recuperació funcional de força i mobilitat amb les dues diferents modalitats. Un altre objectiu que es durà a terme durant l’estudi és avaluar el feedback dels pacients sobre l’ús d’aquestes tecnologies i parlar de possibles millores. Aquest estudi s’ha dut a terme al Departament de Rehabilitació de la seu Plató de l’Hospital Clínic de Barcelona. Totes les fases implicades, des de la selecció dels pacients participants fins a la discussió dels resultats, es detallen en aquest projecte
Pipeline development for realistic synthetic database generation of SPECT neuroimaging
(2025-06-11) Álvarez Gamero, Sara; Tudela Fernández, Raúl
This Final Degree Project introduces a simulation pipeline for generating realistic brain SPECT images using Monte Carlo methods. Built on real anatomical MRI and CT data from patients with diagnosed or suspected Parkinson’s disease, the pipeline creates synthetic images by first generating activity and attenuation maps that replicate radiotracer distribution and tissue densities. These are then used as inputs for the SimSET simulation engine to produce synthetic SPECT projections. A core feature of the pipeline is its iterative framework. Simulated reconstructions are compared to real clinical SPECT images using anatomical atlases to quantify regional differences. These discrepancies inform successive updates to the activity map, gradually refining image realism across iterations. The approach enables the generation of synthetic SPECT studies that closely resemble real data, while maintaining known ground truth—key for evaluating and validating quantification methods in nuclear medicine. The pipeline is modular, reproducible, and scalable, with integrated quality controls and standardized preprocessing steps. It lays the groundwork for creating a database of synthetic realistic neuroimaging studies. The project contributes to the advancement of validation for quantification imaging tools, particularly for Parkinson’s disease research and clinical validation.
MECHANICAL REGULATION OF NUCLEOCYTOPLASMIC TRANSPORT IN NEURONS
(2025-06-11) Altadill Cordero, Mar; Roca-Cusachs Soulere, Pere
La rigidesa del cervell varia amb l’envelliment, i s’observa substancialment alterada en trastorns neurodegeneratius, com per exemple l’Alzheimer. Avui dia, està àmpliament acceptat el fet que les cèl·lules són capaces de detectar i respondre a estímuls mecànics de l’entorn, en un procés conegut com a ‘mechanosensing’. Un dels mecanismes més sensibles a les forces mecàniques és el transport nucleocitoplasmàtic (TNC), on els impulsos mecànics externs són capaços de modular la tensió nuclear de les cèl·lules, com també de causar un impacte sobre el tràfic molecular a través dels porus nuclears. A més a més, s’ha evidenciat que el TNC està alterat en patologies com l’Alzheimer o en altres malalties neurodegeneratives. No obstant això, encara es desconeix si les neurones poden detectar canvis en la rigidesa cerebral, afectant al transport entre el nucli i el citoplasma conseqüentment. Aquest projecte respon aquesta qüestió mitjançant la combinació d’un model neuronal (cèl·lules SH-SY5Y diferenciades en neurones), hidrogels de poliacrilamida (PAA) d’una rigidesa determinada, i un sensor sintètic mecano-sensitiu (SeNCYT) capaç de reportar canvis en el TNC. Per validar els resultats, s’han aplicat xocs osmòtics i s’han comparat els efectes sobre cèl·lules diferenciades i cèl·lules i no diferenciades. Els resultats indiquen que les cèl·lules neuronals són sensibles a la rigidesa del substrat, incloent-hi alteracions en la morfologia cel·lular i en la dinàmica del TNC, fet que suporta la idea que el ‘mechanosensing’ contribueix a la regulació del transport nuclear en patologies neurodegeneratives. Aquestes conclusions permeten enllaçar la mecànica cerebral amb la disfunció neuronal.
KeyCARE: a framework for biomedical Keyword Extraction, term Categorization, and semantic Relation
(2024-06-05) Marsol Torrent, Sergi; Barrios, Juan Ignacio
The medical sector generates vast amounts of unstructured data, which, if processed correctly, can significantly enhance medical processes and their outcomes. This thesis presents the development of KeyCARE, a Python library for keyword extraction, term categorization, and relations extraction that tackles this need. Utilizing mainly unsupervised and few-shot methods, KeyCARE efficiently extracts classified keywords from medical records with a recall of up to 98% and an f-score of up to 61%, with partial overlaps considered as correct. While these scores are not comparable to those of supervised Named Entity Recognition systems, they set a high standard for an unsupervised alternative in scenarios of data scarcity. Moreover, the library incorporates relation extractors that identify hierarchical relationships among biomedical keywords and with terminologies, achieving a precision and recall of 93%. This has a clear application in terminology enrichment, data generation and information extraction, particularly in specific domains and low-resource languages such as Catalan. This thesis encompasses the comprehensive development of KeyCARE, including an in-depth evaluation of the implemented models as well as basic use cases demonstrating its practical applications.
Leveraging single-cell genomics for cell therapy in Huntington's disease
(2024-06-05) Lozano Gargallo, Clàudia; Colomer, Jordi
brain regions,particularlyaffectingthestriatalneuronsknownasMediumSpinyNeurons(MSNs). Unfortunately,neuronsdonotregenerate,thus,makingitimpossibleforthebodytoreplacetheir function. Despiteextensiveresearch,thereiscurrentlynocureforHD,andthegoaloftreatmentis to alleviatesymptoms. Stem celltherapyrepresentsapromisingdirection,allowingindefinite in vitro cells expansionand differentiationintothedesiredtargetneurons.Themainlimitationofneuralcellstransplantisthat once theirprojectionsaregrown,theycannotbedetractedfromthecultureplatewithoutbreaking and causingcelldeath. In thisproject,wefocusonidentifyingtheoptimaltargetprogenitorsubpopulation.Thecellsshould be matureenoughtoexclusivelydifferentiateintoMSNsbutyetnotoverlymature,preventingthe development oftheprojectionsthatleadtocelldeathuponbreaking.Leveragingcomputational genomics, weanalyzehumansingle-cellRNAsequencingdatatocharacterizetheMSNdifferen- tiation process.Integrating in vivo and in vitro data, wegroupcellsindifferentclustersdepending on theirgeneticexpressionandcomputetheirvelocitiestoanalysetheirfate.Ourfindingsreveala potential groupofcellsthatseemtofulfillallthecriteria.Additionally,theprojecthasenabledthe characterization ofthedifferentlineagesofcellswithinthedifferentiationprocessofstriatalneurons.

Incidència: Per tasques de manteniment, el Dipòsit Digital només funciona per a la consulta.

Examinar

Enviaments recents