ESB - Dissertações de Mestrado / Master Dissertations
URI permanente para esta coleção:
Navegar
Entradas recentes
- Exploring the potential of Portuguese thermal spring water as an ingredient for cosmetic formulationsPublication . Rocha, Pedro Emanuel de Almeida; Madureira, Ana Raquel Mendes Ferreira Monteiro de; Pedrosa, Sílvia Marlene Almeida SantosThermal spring waters (TSW), utilized for centuries for their health benefits, are increasingly recognized scientifically for their therapeutic potential. These are largely attributed to their unique mineral compositions derived from long-term water-rock interactions. Properties like immunomodulation, antioxidant activity, and skin barrier reinforcement have spurred interest in TSW within the cosmetics industry, driven by consumer demand for natural and effective ingredients. This thesis aims at investigating the cosmetic potential of TSW from Chaves, Portugal, a historically recognized spring whose properties remained scientifically underexplored. The initial phase characterized Chaves TSW and evaluated its dermatological bioactivities. In vitro studies revealed anti-inflammatory potential, evidenced by reduced IL-6 levels in human keratinocytes (HaCaT) exposed to urban pollutants, alongside with anti-elastase activity. A clinical study on 23 human volunteers demonstrated that Chaves TSW significantly improved skin hydration on dry skin volunteers and reduced transepidermal water loss (TEWL), suggesting skin barrier enhancement without disrupting the resident skin microbiota. Based on these positive skin findings, a cosmetic formulation incorporating Chaves TSW was developed with a high natural index (96% based on ISO 16128). The research scope was extended to evaluate the effects on hair, a less common application for TSW. While Chaves TSW alone did not significantly alter bleached hair fiber’s properties, the formulation improved hair combability, likely due to included surfactants. Conversely, the formulation decreased the thermal denaturation temperature of hair keratin, potentially linked to surfactant-keratin interactions observed via DSC and ATR-FTIR analyses. Finally, the base formulation's potential was further explored by incorporating Mallow (Malva sylvestris); endemic to the region; and Cucumber (Cucumis sativus) extracts, known for their skin benefits, such as antioxidant potential. Three formulations containing both extracts and a mixture of both were developed, and subsequent testing revealed significantly improved bioactivities. Notably, a strong synergistic Fe2+ chelating capacity emerged, exceeding that of the formulation's chelating agent alone. This correlated with enhanced antioxidant activity and significantly increased inhibition of the age-related enzymes collagenase and tyrosinase, likely due to cofactor chelation. The combination of extracts also yielded potent elastase inhibition. Furthermore, the anti-inflammatory effects were re-evaluated using a mixture containing TSW extracts, and sodium phytate (chelating agent), which nonetheless exhibited anti-inflammatory properties like those of water but effectively reducing IL-1α levels. In conclusion, this work validates the beneficial properties of Chaves TSW for skincare, particularly its anti-inflammatory and barrier-reinforcing effects. It further demonstrates that formulating TSW, especially when enhanced with synergistic natural extracts like Mallow and Cucumber, can create multifunctional cosmetic ingredients with significantly amplified antioxidant, anti-enzyme, and anti-inflammatory activities, offering promise for advanced skin and potentially scalp care applications.
- Potential bacterial carriers of antibiotic resistance in aquaculture and aquaponicsPublication . Amado, Nefissa Carlie; Moreira, Ivone Cristina Vaz; Rodrigues, Célia Maria ManaiaAquaculture (aquatic animal farming) and Aquaponics (an integration of aquaculture and soilless plant cultivation) systems are increasingly used to simultaneously produce fish and plants within the same circulating water system. Fish skin mucus is in direct contact with the environment, functioning as a physical, biochemical, and semipermeable barrier for the fish. Moreover, the bacterial communities within fish skin mucus may reflect the bacterial diversity of the surrounding water and consequently of the resistome (group of genes associated to antibiotic resistance). Fish skin mucus is also described as having associated antimicrobial activity. This dissertation aimed to assess whether fish and plants cultivated in aquaculture/aquaponics systems can be colonized by antibiotic resistant bacteria and antibiotic resistance genes, and if fish skin mucus are more prone to accumulate antibiotic resistance than aquaculture biofilms. For that, both culture dependent and culture independent methods were used to study the occurrence of antibiotic resistant bacteria and genes in samples from the fish skin mucus, biofilms of the aquaculture tanks and in plants growing in an experimental aquaponic system. Bacterial growth was assessed on GSP agar (selective for Pseudomonads and Aeromonads) and GSP agar supplemented with cefotaxime (4 mg/L) and streptomycin (50 mg/L). Isolates (n=114) were selected for phenotypic antibiotic resistance profiling and genotypic analysis targeting resistance genes for β-lactams (blaCTX, blaSHV, and blaOXA), and genes associated to mobile genetic elements (intI1 and incF). Bacterial isolates (n=22) harboring the intI1 gene were further analyzed for the presence of class 1 integrons and integron variable regions sequenced. Additionally, qPCR was used for the quantification of the 16S rRNA gene as well as the intI1 integrase gene, a biomarker for antibiotic resistance, in samples of water and lettuce roots from the experimental aquaponics system. The highest bacterial growth occurred on GSP without selective pressure, while the addition of streptomycin to the GSP culture medium resulted in the most inhibitory effect on the growth of bacteria especially in the biofilm and roots samples (<1.4% of tolerant bacteria). Overall, 3.6-38.6% (average values) of the cultivable bacteria growing on GSP tolerated the 4 mg/L of cefotaxime while just 0.2-3.9% (average values) tolerated the 50 mg/L of streptomycin. Most of the bacterial isolates recovered were Gram-negative (111/114), catalase positive and cytochrome c oxidase negative. When observing the antibiograms results, resistance to cephalothin and streptomycin were among the most prevalent (60-98% and 34-70%, respectively) in the isolates of the different origins. In contrast isolates showed a low prevalence of resistance to tetracycline, meropenem, ciprofloxacin or colistin (<15%). Multi drug resistances were mainly observed in isolates recovered from biofilms and roots. Fish skin mucus isolates were identified to be mainly Acinetobacter johnsonii, Achromobacter spp., and Flavobacterium spp. while the isolates recovered from the biofilms present in the same aquaculture system were predominantly identified as E. coli and Achromobacter spp. The main species identified in the lettuce roots were Pseudomonas monteilii and Enterobacter ludwigii. Genes associated with mobile elements intI1 and incF were detected in <20% of isolates mostly E. coli from biofilms. The blaSHV gene was just detected in E. coli isolates from biofilm, for which were also detected the genes intI1 and incF. Class 1 integrons were detected in 18 isolates mostly from biofilms, with one present in a biofilm E. coli isolate, confirmed to carry the aminoglycoside resistance gene aadA1. These results demonstrate that biofilms are a major hotspot for bacteria and genes associated with antibiotic resistance, while fish skin mucus and lettuce roots were not identified as important hotspots.
- CardioRiskAI : a clinical application for rapid and affordable cardiovascular risk assessment using ECG analysis and machine learningPublication . Silva, Martim José de Pinho Gutierres; Rodrigues, Pedro Miguel de Luís; Guimarães, Vânia Margarida CardosoCardiovascular diseases (CVDs) are the leading global cause of death, yet traditional risk assessment is often slow and resource-intensive. This work investigates a novel approach for rapid, accessible CVD risk assessment using engineered electrocardiography (ECG) features combined with minimal clinical data (age, BMI, smoking, and diabetes status). By leveraging increasingly portable and inexpensive ECG technology, we aim to provide a cost-effective alternative to lab-dependent methods like the Framingham score. An ECG processing pipeline was developed to extract 102 features, which were used to train classification and regression models targeting the Framingham risk score. The models demonstrated excellent performance: binary classification distinguishing between low- and high-risk patients achieved 97.4% accuracy (AUC 0.99); the multiclass comparison among low-, moderate-, and high-risk groups reached 86.79% accuracy (AUC = 0.88); and regression analysis produced RMSE of between 4 to 6 percentage points. To translate these findings into practice, a clinical application was co-developed with cardiologists, integrating secure authentication, patient management, explainable AI, and automated reporting. The results indicate that a compact set of ECG-derived features, coupled with basic clinical information, can provide rapid and cost-effective cardiovascular risk assessment. This approach holds the potential to democratize risk assessment in low-resource settings and accelerate urgent care evaluations. If validated for generalizability, these models could significantly reduce the reliance on extensive and costly blood laboratory tests for cardiovascular risk stratification.
- Quantitative sensor for glucose monitoring in pet's salivaPublication . Aroso, Maria Francisca Paes Cardoso de Lacerda; Mesquita, Raquel Beatriz Ribeiro de; Rangel, António Osmaro Santos SilvaDiabetes mellitus is a common endocrine disease in dogs and cats, with diagnosis and monitoring traditionally relying on invasive blood sampling, often associated with stress, discomfort, and inconsistent results. Saliva has emerged as a promising non-invasive alternative for detecting biomarkers such as glucose. This work aimed to develop and optimize a paper-based microfluidic device (µPAD) for quantitative glucose detection in dogs' and cats' saliva. The sensor was designed using layered paper discs loaded with enzymatic reagents (glucose oxidase and peroxidase) and o-dianisidine as a chromogenic substrate, generating a colourimetric response directly proportional to glucose concentration. Optimization studies focused on paper type, sample volume, and enzyme concentration, resulting in a detection limit of 6.7 mg/L and a dynamic range of 22.0 – 220 mg/L, covering both physiological and pathological glucose concentrations in dogs. The device showed good reproducibility (RSD = 4%) and low fabrication costs (< €5 per unit). Tests with canine saliva confirmed the ability to detect glucose, although limited by the reduced number and volume of samples available. Overall, the developed µPAD demonstrates strong potential as a non-invasive tool for monitoring glycemia in pets, supporting more accessible, affordable, and animal-friendly diabetes management.
- Early detection of Alzheimer's disease using MRI and 2D machine learning systems : enhancing clinical diagnosticsPublication . Menezes, José Maria Duarte de Magalhães e; Rodrigues, Pedro Miguel de Luís; Barbosa, Maria Inês AraújoThe early detection of Alzheimer’s disease (AD) is critical for effective intervention and management. However, traditional classification systems—typically limited to cognitively normal (CN), mild cognitive impairment (MCI), and AD stages—often fail to capture the clinical heterogeneity and progressive nature of cognitive decline. This study introduces a comprehensive machine learning (ML) framework to enhance AD classification using structural Magnetic Resonance Imaging (MRI). We implement a refined nine-stage system that not only subdivides MCI into Early (EMCI), transitional (MCI), and Late (LMCI) stages but also incorporates beta-amyloid biomarker status (positive/negative) for further differentiation. Our research analyzed T1-weighted MRI scans from 714 subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database, categorized into nine groups: AD, CN, Subjective Memory Complaints (SMC), and amyloid-stratified versions of EMCI, MCI, and LMCI. A robust preprocessing pipeline was applied, followed by feature extraction using Histogram and Gray-Level Co-occurrence Matrix (GLCM) methods. Pairwise classification was performed using a suite of ML classifiers belonging to Scikit-learn library and feature selection techniques. The framework demonstrated high classification accuracy across numerous stage comparisons, achieving standout results such as 97.8% for AD vs. EMCI-Negative and 95.5% for AD vs. CN. Notably, the model also effectively distinguished between adjacent MCI stages, like MCI-Negative vs. LMCI-Negative, with 94.9% accuracy. Our findings indicate that beta-amyloid status enhances discriminability and that Histogram and GLCM features provide complementary diagnostic information, with the coronal plane frequently yielding the most informative results. This multi-stage classification approach shows significant promise for improving the precision of early and differential AD diagnosis.
- Assessing banana and rice waste for postbiotics productionPublication . Carvalhas, Joana Maria Santiago Jales; Madureira, Ana Raquel Mendes Ferreira MonteiroThe growing demand for sustainable solutions in biotechnology has encouraged the use of agri-food residues as alternative raw materials for producing value-added compounds. Accordingly, this study aimed to propose a sustainable biotechnological strategy for valorizing residues, focusing on banana peel and rice bran, through the microbial production of health-promoting compounds. In line with the principles of the circular economy, the work sought to demonstrate that food by-products can be converted into bioactive compounds, thereby combining environmental responsibility with applications relevant to the health sector. Initially, both residues were characterized to evaluate their suitability for bioprocessing. Although an attempt to extract ellagic acid from banana peel using ultrasound-assisted extraction (UAE) was performed, the compound was not detected. Consequently, the investigation shifted towards the microbial production of γ-aminobutyric acid (GABA). To this end, the residues were subjected to enzymatic hydrolysis with two proteases, alcalase and P24, to increase the availability of glutamic acid (Glu). Subsequently, the resulting hydrolysates were fermented with co-cultures of Saccharomyces cerevisiae and either Lactobacillus brevis or Lactobacillus plantarum. Samples were collected at 0, 48 and 72 hours for GABA and Glu quantification. The results indicated that rice bran was a more suitable substrate than banana peel for GABA production, primarily because of its higher protein content and greater release of Glu after hydrolysis with P24. Moreover, the co-culture with L. brevis yielded higher GABA levels at earlier fermentation stages, whereas L. plantarum displayed a slower and more gradual profile. This contrast highlights distinct fermentation dynamics and suggests different strategies for process optimization depending on the microbial consortium and production goals. Overall, this study demonstrates the potential of fermentation-based approaches for the valorization of agro-industrial residues. Not only did the process prove effective in generating GABA from rice bran, but it also pointed to possible strategies for improving the use of banana peel. Therefore, future studies should focus on optimizing fermentation conditions to enhance both efficiency and applicability.
- Quality assessment and anti-listerial efficacy of cold plasma treatment on fresh lettucePublication . Ersoylu, Ekin; Brandão, Teresa Maria Ribeiro da Silva; Barbosa, Joana Cristina PachecoThis work evaluates the potential of cold plasma technology as a decontamination method for fresh lettuce. Specifically, it investigates the effectiveness of cold plasma in reducing Listeria innocua (a non-pathogenic surrogate for L. monocytogenes) inoculated onto lettuce leaves, while simultaneously assessing its impact on quality attributes, including color (total color difference in relation to fresh samples, chroma, and browning index), pH, chlorophylls (a, b, and total), and total phenolic content. A conventional hot water treatment (blanching at 80 °C) was used as reference. By examining both microbial inactivation efficiency and preservation of physicochemical properties, the study aims to determine whether cold plasma can serve as a reliable approach for ensuring lettuce decontamination without compromising quality. Cold plasma treatments (gliding arc discharge) with exposure times of 5, 10, 15, 20, 25, and 30 s were compared with thermal treatments of equal duration. pH values remained relatively stable (5.636.0), though thermal samples were more uniform, while plasma samples showed slightly greater variability. Color analysis showed treatment-dependent effects: plasma preserved color and vividness more effectively (higher chroma values), while thermal treatment better limited browning. Chlorophyll a remained relatively stable under both treatments, with no significant differences across most time points. Chlorophyll b was more sensitive: at 5 and 30 s, plasma-treated samples retained significantly higher levels than thermal, suggesting superior preservation of pigments. Total phenolic content also differed markedly between treatments. Plasma-treated lettuce maintained or slightly increased phenolic levels over time (j14315 mg GAE/g FW), while thermal treatment caused significant reductions from 5 s onward, with final values more than two-fold lower than plasma. Regarding microbial inactivation, thermal treatment completely eliminated L. innocua within 5 s (~7-log reduction), while cold plasma produced a slower decline, achieving only ~2-log reduction after 30 s. Despite its lower inactivation efficiency, cold plasma better preserved color, chlorophyll b, total chlorophyll, and phenolic content compared to thermal processing. Therefore, while thermal treatment ensures decontamination, cold plasma represents a promising non-thermal preservation strategy for lettuce with low microbial loads, provided its antimicrobial efficacy can be further optimized without compromising quality and freshness.
- Enhancing shelf-life of live clams via modified atmosphere combined with physical confinement through packagingPublication . Goes, Cintia Borghetti; Poças, Maria de Fátima dos Reis Filipe Tavares; Vieira, Tiago Daniel MonteiroClams are the most expensive seafood products because they have a low yield compared to other bivalve’s molluscs. Venerupis corrugata (pullet carpet shell) is a premium European native clams’ species due to high gastronomic, nutritional and economic value, found and appreciated in Portugal. This species is highly perishable. It is typically sold alive within three or four days under refrigeration. The shelf-life period is defined based on the time the clams are alive. Significant risk of loss and waste exists during this period and efforts to extend its shelf-life are relevant. This study aimed to evaluate the shelf-life and physiological quality of live V. corrugata clams stored at 3 ± 1 °C in different packaging systems evaluating the effect of high-oxygen modified atmosphere packaging (MAP) versus ambient air, and of physical confinement in net bags versus loose storage. Survival percentage, gases concentration, volatile organic compounds (VOCs), pH, glycogen content, and biogenic amines were monitored throughout storage (7 days). Results showed that survival and physiological quality were strongly influenced by the combined effects of oxygen availability and confinement. High-oxygen MAP delayed mortality by sustaining aerobic metabolism for longer, thereby reducing the respiration quotient. Physical confinement was critical in maintaining intravalvular liquid and preserving clam viability. Nevertheless, the modelled survivability extension was limited to only 1 day under optimal MAP conditions and confinement, highlighting the intrinsic sensitivity of this species.
- The resilience of listeria monocytogenes to environmental stressorsPublication . Gonçalves, Catarina Ferreira; Magalhães, Rui Miguel Barros de Sousa; Teixeira, Paula Cristina MaiaThe protection of public health requires the implementation of rigorous food safety measures, particularly in relation to ready-to-eat products. Contamination by foodborne pathogens remains a major concern, especially in food-processing environments where microbial persistence can compromise hygiene standards. Among these pathogens, Listeria monocytogenes stands out due to its remarkable ability to survive under adverse conditions such as desiccation and high salinity. These traits complicate sanitation practices and increase the risk of contamination, especially when hygiene protocols in food processing environments are inadequate. Its ability to cross physiological barriers further reinforces its significance as a public health threat. In this context, the present study investigated the survival and stress responses of persistent, non-persistent, and clinical strains of L. monocytogenes under environmental conditions commonly associated with food products and processing facilities. The main objective was to advance the understanding of the bacterium’s adaptive responses, with implications for food safety and public health, by analyzing strain behavior under stress conditions. The specific objectives were: (i) to assess survival dynamics in sodium chloride at 20 °C; (ii) to evaluate survival following desiccation on stainless steel at 10 °C; and (iii) to determine the minimum inhibitory concentrations (MICs) of strains for benzalkonium chloride and peracetic acid. The results showed that persistent and non-persistent strains exhibited similar survival dynamics under osmotic and desiccation stress, with no statistically significant differences in inactivation kinetics. In contrast, disinfectant susceptibility assays revealed marked variability: while persistent and non-persistent strains demonstrated comparable responses to benzalkonium chloride, clinical isolates displayed significantly higher MIC values, indicating reduced sensitivity. For peracetic acid, susceptibility levels were similar across groups, although heterogeneity among individual strains indicated distinct adaptive responses. This study provides valuable insights into the survival and stress tolerance of L. monocytogenes under conditions relevant to food-processing environments. The findings reinforce the need for robust and comprehensive hygiene strategies, capable of addressing strain-level variability, thereby strengthening food safety management and public health protection.
- Identification of food product development opportunities from goat's cheese production by-products at a factory in FundãoPublication . Oladipo, Ayomidotun Abigail; Pintado, Maria Manuela Estevez; Peça, João Diogo AndréThis study was conducted within the framework of the Centre for Food Education and Research (CFER) to promote the valorization of food industry by-products through research and innovation. It focused on analyzing and comparing the nutritional composition of by products from goat cheese production in both laboratory and industrial settings, to assess their reproducibility and potential for food product development. In collaboration with a local cheese factory in Fundão, the work aimed to identify innovative ways to utilize goat’s milk whey and transform it into value-added food products either for direct consumption or as an ingredient in food production. Whey collected from an artisanal factory in Fundão was used to produce Ricotta and Travia in a laboratory by replicating methods and conditions used in the factory. The samples were chemically analyzed for protein, fat, ash, carbohydrate, moisture, and total solids content and statistically compared to industrial samples using ANOVA and t-tests. Ricotta and Travia samples developed in the laboratory consistently showed a markedly greater protein and lower fat content than their industrial counterparts. Moisture and total solids showed no notable differences (p > 0.05), whereas protein, fat, ash, and carbohydrate levels differed significantly (p < 0.05). These results validate the possibility and effectiveness of laboratory recreation with nutritional attributes, emphasizing the opportunity to develop functional foods and encouraging the sustainable utilization of dairy by-products.