https://rd.uffs.edu.br/handle/prefix/67 Mon, 31 Aug 2026 03:35:14 GMT 2026-08-31T03:35:14Z https://rd.uffs.edu.br/handle/prefix/9407 Title: A aplicação de smart contracts para o compartilhamento de dados em healthcare Author: Barbosa, Rian Borges First advisor: Schreiner, Geomar André Abstract: In recent years, the medical field has relied on a centralized database, a repository for unifying health data such as patient records. However, this proposal presents systemic security challenges, as it involves confidential information. Unifying this data through a decentralized network would address these issues and facilitate large-scale data man agement. This work proposes the use of smart contracts in conjunction with blockchain technology for the centralization and sharing of health data. The developed application proved to be viable, demonstrating that it is possible to ensure data integrity and control access to the information. Publisher: Universidade Federal da Fronteira Sul Type: Monografia Thu, 01 Jan 2026 00:00:00 GMT https://rd.uffs.edu.br/handle/prefix/9407 2026-01-01T00:00:00Z https://rd.uffs.edu.br/handle/prefix/9406 Title: Predição de infarto agudo do miocárdio e insuficiência cardíaca com aprendizado de máquina - uma análise comparativa no MIMIC-IV Author: Bortoli, Luan First advisor: Salton, Giancarlo Dondoni Abstract: Cardiovascular diseases are among the leading causes of morbidity and mortality, particularly acute myo cardial infarction and heart failure, whose clinical severity calls for early identification of at-risk patients. This study developed and compared four supervised classifiers (Logistic Regression, Support Vector Machine, Decision Tree, and K-NearestNeighbors)topredictCVDusingMIMIC-IVdata. Atotalof546,028hospitaladmissionswereanalyzed,witha binary outcome defined by ICD-9/ICD-10 codes; 16.3% were positive cases. Features comprised demographic variables, vital signs, and laboratory tests; preprocessing included removal of physiologically implausible values, median imputa tion, standardization, and a stratified split into training (70%) and testing (30%). Models were tuned via grid search with stratified cross-validation and assessed using accuracy, precision, recall, F1-score, and AUC-ROC, complemented by con fusion matrices and interpretability analyses. The decision tree achieved the best overall balance (AUC-ROC 0.868; recall 0.788; lowest false negatives, 5,664), while the SVM was competitive at higher computational cost. Logistic regression provided stable performance with coefficient-based interpretability, and KNN, despite the highest accuracy, showed low recall, underscoring the limits of accuracy under class imbalance. Interpretability results highlighted age, urea, troponin T, and NT-proBNP as the most influential variables, supporting the clinical plausibility of the learned patterns. Publisher: Universidade Federal da Fronteira Sul Type: Monografia Thu, 01 Jan 2026 00:00:00 GMT https://rd.uffs.edu.br/handle/prefix/9406 2026-01-01T00:00:00Z https://rd.uffs.edu.br/handle/prefix/9405 Title: Resolução do problema de alocação de bombeiros comunitários para o corpo de bombeiros militar de santa catarina– 6º BBM, Chapecó Author: Silva, Arthur Emanuel da First advisor: Braga, Andrei de Almeida Sampaio Abstract: The development of work schedules for community firefighters represents a major chal lenge for the Military Fire Department of Santa Catarina (CBMSC), due to the need to balance volunteer availability, minimum shift coverage, and equitable assignment distribution. This work proposes an Integer Linear Programming (ILP) model for the automatic generation of these sche dules and comparison with the manual approach. The model was developed based on requirements gathered with the institution and evaluated using real availability data and manual schedules cove ring fifteen months. The results indicate that the model outperformed the manual approach in most evaluated metrics, increasing shift coverage, reducing operational constraint violations, and ensu ring a more equitable distribution of assignments. Furthermore, schedules were generated within seconds, highlighting the potential of the proposed approach to optimize operational planning in emergency services. Publisher: Universidade Federal da Fronteira Sul Type: Monografia Thu, 01 Jan 2026 00:00:00 GMT https://rd.uffs.edu.br/handle/prefix/9405 2026-01-01T00:00:00Z https://rd.uffs.edu.br/handle/prefix/9404 Title: Análise e solução de problemas propostos na maratona de programação SBC dos anos 2023, 2024 e 2025 Author: Balestrin, Marco Antonio First advisor: Braga, Andrei de Almeida Sampaio Abstract: The present work aims to promote an analysis and discussion of problem solutions from different editions of the SBC Programming Marathon. To this end, a selection of problems was made, taking into account each problem’s difficulty and theme. Then, for each problem, a C++ solution was implemented, with its time and memory com plexity verified, in addition to being evaluated through an online judge. The results obtained reflect the effectiveness of the solutions, based on the following concepts: frequency arrays for block processing, bitwise operations for number manipulation, depth-first search to identify specific cycles in a graph, and modular exponentiation to compute modular multiplicative inverses. In conclusion, by presenting detailed solu tions for different computing topics and explaining the structures and algorithms used, this work significantly contributes to the dissemination of study material aimed for fu ture programming marathon competitors, as well as students interested in competitive programming. Publisher: Universidade Federal da Fronteira Sul Type: Monografia Thu, 01 Jan 2026 00:00:00 GMT https://rd.uffs.edu.br/handle/prefix/9404 2026-01-01T00:00:00Z