RUN
Repositório Institucional da Universidade NOVA de Lisboa
Entradas recentes
Risk Phenotyping Before Graft Implantation
Publication . Ramalhete, Luis; Araújo, Rúben; Vigia, Emanuel; Vieira, Miguel Bigotte; Ferreira, Anibal; Calado, Cecilia R.C.; NOVA Medical School|Faculdade de Ciências Médicas (NMS|FCM); iNOVA4Health - pólo NMS; MDPI - Multidisciplinary Digital Publishing Institute
Background: Rejection remains a major barrier to long-term kidney allograft survival, and pre-transplant risk stratification remains incomplete. This study evaluated whether pre-transplant serum Fourier-transform infrared (FTIR) spectra, analyzed using machine learning methods, could identify kidney transplant recipients at increased risk of subsequent biopsy-proven rejection. Methods: In this retrospective single-center study, 80 pre-transplant serum samples collected on the day of transplantation were initially evaluated; after spectral quality control, 79 samples were retained for analysis. FTIR spectra were acquired in transmission mode and analyzed in the 600–1900 cm−1 and 2800–3400 cm−1 regions. Multiple preprocessing strategies were assessed, including Rubber Band baseline correction, vector normalization, and first- and second-derivative transformation, with and without normalization. Naïve Bayes classifiers with Leave-One-Out Cross-Validation and Fast Correlation-Based Filter feature selection were applied. Results: Exploratory analysis showed broad overlap between groups, indicating a subtle multivariate spectral signal. In the initial exploratory workflow, classifier performance depended strongly on preprocessing and feature selection. Because non-nested feature selection may produce optimistic estimates, the main supervised analysis was repeated using FCBF nested within each LOOCV training fold. The best-performing nested model was obtained using second derivative transformation followed by normalization in the combined 600–1900 and 2800–3400 cm−1 regions, achieving an AUC of 0.837, accuracy of 0.747, sensitivity of 0.675, specificity of 0.821, balanced accuracy of 0.748, and F1-score of 0.730. Permutation testing with 1000 label-randomized repetitions supported performance above chance expectation, with no permuted model reaching the observed AUC (empirical p = 0.000999). Conclusions: Pre-transplant serum FTIR spectroscopy combined with leakage-aware nested machine learning analysis identified an internally validated spectral signal associated with subsequent biopsy-proven rejection. These findings support FTIR as a promising complementary and hypothesis-generating approach for pre-transplant biochemical risk phenotyping, requiring external multicenter validation before clinical application.
Primary Ciliary Dyskinesia
Publication . Teixeira-Oliveira, Sofia; Pais-Cunha, Inês; Lopes, Susana S.; Seixas, Susana; Ferraz, Catarina; Amorim, Adelina; Azevedo, Inês; NOVA Medical School|Faculdade de Ciências Médicas (NMS|FCM); Elsevier Espana S.L.U
Introduction: Primary Ciliary Dyskinesia (PCD) is a rare genetic disorder caused by defective ciliary structure and function, leading to chronic respiratory and systemic manifestations. Diagnostic pathways have evolved over time, but no single standalone test exists. Methods: Retrospective study of PCD patients followed at a Portuguese tertiary hospital, between 2001–2024. Results: Thirty-five patients with a confirmed diagnosis were included: 13 children and 22 adults. Median age at diagnosis was 7 years (0–16) in children and 38.5 years (12–64) in adults. Time to diagnosis decreased over the years, coinciding with a shift in the hierarchy of methods from high-speed video microscopy and transmission electron microscopy to genetic testing. The most frequent mutations were DNAH5 (31.3%) and DNAH11 (12.5%). Pulmonary function tended to be better in children (p = 0.085), whereas bronchiectasis were more extensive and bilateral in adults (p = 0.044 and p = 0.002, respectively). Children more frequently received treatment with hypertonic saline (p = 0.003) and adults with bronchodilators (p = 0.035). Pseudomonas aeruginosa was only identified in adults; inhaled antibiotics were only prescribed in this age group (18.2%). Conclusion: This represents the largest Portuguese cohort to date and provides relevant clinical and diagnostic insight into age-related differences, supporting the importance of early detection and intervention to limit lung damage.
Controversial Issues as a catalyst for voluntary oral production
Publication . Pestana, João Filipe Banha; Prakash, Rima Jay
This aimed to observe the impact of controversial issues in the voluntary oral production of highschool students. My research involved creating classes centered around contextually relevant controversial issues in order to observe if they had any positive impact in the voluntary oral production of students. Their voluntary oral production was tallied and compared in order to observe relevant fluctuations, with the intent of drawing conclusions towards the main research question of this report: Can controversial issues be a catalyst for voluntary oral production.
NextGenerationEU landing
Publication . Scotti, Francesco; Caporali, Carlo; Luca, Davide; NOVA School of Business and Economics (NOVA SBE); Wiley
In response to the COVID-19 crisis, the European Union introduced NextGenerationEU, its largest stimulus package to date. We focus on Italy, the program's largest beneficiary, and analyse the territorial distribution of funds. Although centrally coordinated, most expenditures were awarded through calls and implemented by local governments. Using a sample of approximately 4500 Italian municipalities for which administrative-efficiency measures can be constructed, we estimate a two-stage Heckman selection model to distinguish between realized access to funding and award intensity conditional on participation. Results show that funds are disproportionately allocated to Southern, relatively poorer urban municipalities with stronger administrative capacity and prior experience with EU Cohesion Policy. We then test if, across different policy areas, funds were targeted following specialization or convergence logics. We find that, across some of the key policy areas, such as digital, education and healthcare, the allocation is reinforcing existing spatial specialization, as opposed to targeting “lagging behind” territories. Finally, we uncover how administrative efficiency and stronger local governance are significant predictors of reduced project delays, while we do not find a significant North-South divide here. Taken together, these findings highlight key allocative trade-offs and the uneven administrative capacity of territories expected to implement the program.
An AI-Enhanced Bibliometric Analysis of Customer Feedback and Hotel Performance in Hospitality
Publication . Neves, Marta Morgado; António, Nuno; Moro, Sérgio; NOVA Information Management School (NOVA IMS); Information Management Research Center (MagIC) - NOVA Information Management School; Universidade do Algarve, School of Management, Hospitality and Tourism
In the digital era, customer feedback has evolved into a strategic asset influencing hotel performance, competitiveness, and investment decisions. This study reviews the relationship between online reviews and key performance indicators (KPIs) in the hospitality sector and proposes an AI-enhanced literature review framework. By integrating bibliometric analysis, Natural Language Processing (NLP), and Large Language Models (LLMs), the approach enables scalable, context-aware synthesis of research on customer perceptions and their financial implications. The findings show that customer feedback significantly influences hotel performance, particularly affecting pricing strategies, occupancy rates, and revenue outcomes. However, the literature remains heavily reliant on rating-based proxies and makes limited use of real operational data, thereby constraining the accuracy of financial assessments. Additionally, gaps persist in data integration, geographic representativeness, and the translation of feedback into actionable investment decisions. From a practical perspective, the results highlight the need for more advanced, data-driven frameworks that integrate textual feedback with financial metrics. The proposed methodology offers a replicable approach to support more accurate and strategically relevant analyses in hospitality management.