Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Larissa Montenegro has a total of 14 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    What We Know About the Role of Large Language Models for Medical Synthetic Dataset Generation

    by Larissa Montenegro, Luis M. Gomes, José M. Machado

    Synthetic medical text generation has emerged as a solution to data scarcity and privacy constraints in clinical NLP. This review systematically evaluates the use of Large Language Models (LLMs) for structured medical text generation, examining techniques such as retrieval-augmented generation (RAG), structured fine-tuning, and domain-specific adaptation. Four search queries were applied following the PRISMA methodology to identify and extract data from 153 studies. Key benchmarking metrics, such as performance measures, and qualitative insights, including methodological trends and challenges, were documented. The results show that while LLM-generated text improves fluency, hallucinations and factual inconsistencies persist. Structured consultation models, such as SOAP and Calgary–Cambridge, enhance coherence but do not fully prevent errors. Hybrid techniques that combine retrieval-based grounding with domain-specific fine-tuning improve factual accuracy and task performance. Conventional evaluation metrics (e.g., ROUGE, BLEU) are insufficient for medical validation, highlighting the need for domain-specific benchmarks. Privacy-preserving strategies, including differential privacy and PHI de-identification, support regulatory compliance but may reduce linguistic quality. These findings are relevant for clinical NLP applications, such as AI-powered scribe systems, where structured synthetic datasets can improve transcription accuracy and documentation reliability. The conclusions highlight the need for balanced approaches that integrate medical structure, factual control, and privacy to enhance the usability of synthetic medical text.


  • Behaviour of Machine Learning algorithms in the classification of energy consumption in school buildings

    by José Machado, António Chaves, Larissa Montenegro, Carlos Alves, Dalila Durães, Ricardo Machado, Paulo Novais

    The significance of energy efficiency in the development of smart cities cannot be overstated. It is essential to have a clear understanding of the current energy consumption (EC) patterns in both public and private buildings. One way to achieve this is by employing machine learning classification algorithms, which offer a broader perspective on the factors influencing EC. These algorithms can be applied to real data from databases, making them valuable tools for smart city applications. In this paper, our focus is specifically on the EC of public schools in a Portuguese city, as this plays a crucial role in designing a Smart City. By utilizing a comprehensive dataset on school EC, we thoroughly evaluate multiple ML algorithms. The objective is to identify the most effective algorithm for classifying average EC patterns. The outcomes of this study hold significant value for school administrators and facility managers. By leveraging the predictions generated from the selected algorithm, they can optimize energy usage and, consequently, reduce costs. The use of a comprehensive dataset ensures the reliability and accuracy of our evaluations of various ML algorithms for EC classification.


  • Open Access

    Revolutionising the Quality of Life: The Role of Real-Time Sensing in Smart Cities

    by Rui Miranda, Carlos Alves, Regina Sousa, António Chaves, Larissa Montenegro, Hugo Peixoto, Dalila Durães, Ricardo Machado, António Abelha, Paulo Novais, et al.

    To further evolve urban quality of life, this paper explores the potential of crowdsensing and crowdsourcing in the context of smart cities. To aid urban planners and residents in understanding the nuances of day-to-day urban dynamics, we actively pursue the improvement of data visualisation tools that can adapt to changing conditions. An architecture was created and implemented that ensures secure and easy connectivity between various sources, such as a network of Internet of Things (IoT) devices, to merge with crowdsensing data and use them efficiently. In addition, we expanded the scope of our study to include the development of mobile and online applications, emphasizing the integration of autonomous and geo-surveillance. The main findings highlight the importance of sensor data in urban knowledge. Their incorporation via Tepresentational State Transfer (REST) Application Programming Interface (APIs) improves data access and informed decision-making, and dynamic data visualisation provides better insights. The geofencing of the application encourages community participation in urban planning and resource allocation, supporting sustainable urban innovation.


  • Intelligent Systems in Healthcare: An Architecture Proposal

    by António Chaves, Larissa Montenegro, Hugo Peixoto, António Abelha, Luís Gomes, José Machado


  • Open Access

    Human-Assisted vs. Deep Learning Feature Extraction: An Evaluation of ECG Features Extraction Methods for Arrhythmia Classification Using Machine Learning

    by Larissa Montenegro, Mariana Abreu, Ana Fred, Jose M. Machado

    The success of arrhythmia classification tasks with Machine Learning (ML) algorithms is based on the handcrafting extraction of features from Electrocardiography (ECG) signals. However, feature extraction is a time-consuming trial-and-error approach. Deep Neural Network (DNN) algorithms bypass the process of handcrafting feature extraction since the algorithm extracts the features automatically in their hidden layers. However, it is important to have access to a balanced dataset for algorithm training. In this exploratory research study, we will compare the evaluation metrics among Convolutional Neural Networks (1D-CNN) and Support Vector Machines (SVM) using a dataset based on the merged public ECG signals database TNMG and CINC17 databases. Results: Both algorithms showed good performance using the new, merged ECG database. For evaluation metrics, the 1D-CNN algorithm has a precision of 93.04%, an accuracy of 93.07%, a recall of 93.20%, and an F1-score of 93.05%. The SVM classifier (λ = 10, C = 10 × 109) achieved the best classification metrics with two combined, handcrafted feature extraction methods: Wavelet transforms and R-peak Interval features, which achieved an overall precision of 89.04%, accuracy of 92.00%, recall of 94.20%, and F1-score of 91.54%. As an unique input feature and SVM (λ=10,C=100), wavelet transforms achieved precision, accuracy, recall, and F1-score metrics of 86.15%, 85.33%, 81.16%, and 83.58%. Conclusion: Researchers face a challenge in finding a broad dataset to evaluate ML models. One way to solve this problem, especially for deep learning models, is to combine several public datasets to increase the amount of data. The SVM and 1D-CNN algorithms showed positive results with the merge of databases, showing similar F1-score, precision, and recall during arrhythmia classification. Despite the favorable results for both of them, it should be considered that in the SVM, feature selection is a time-consuming trial-and-error process; meanwhile, CNN algorithms can reduce the workload significantly. The disadvantage of CNN algorithms is that it has a higher computational processing cost; moreover, in the absence of access to powerful computational processing, the SVM can be a reliable solution.