From Knowledge Extraction to Assertive Response: An LLM Chatbot for Information Retrieval in Telemedicine Systems Bruna D. Pupo, Daniel G. Costa, Roger Immich, Aldo von Wangenheim, Alex Sandro Roschildt Pinto, et al. Applied Sciences Switzerland, 2025 The development of new technologies, improved by advances in artificial intelligence, has enabled the creation of a new generation of applications in different scenarios. In medical systems, adopting AI-driven solutions has brought new possibilities, but their effective impacts still need further investigation. In this context, a chatbot prototype trained with large language models (LLMs) was developed using data from the Santa Catarina Telemedicine and Telehealth System (STT) Dermatology module. The system adapts Llama 3 8B via supervised Fine-tuning with QLoRA on a proprietary, domain-specific dataset (33 input-output pairs). Although it achieved 100% Fluency and 89.74% Coherence, Factual Correctness remained low (43.59%), highlighting the limitations of training LLMs on small datasets. In addition to G-Eval metrics, we conducted expert human validation, encompassing both quantitative and qualitative aspects. This low factual score indicates that a retrieval-augmented generation (RAG) mechanism is essential for robust information retrieval, which we outline as a primary direction for future work. This approach enabled a more in-depth analysis of a real-world telemedicine environment, highlighting both the practical challenges and the benefits of implementing LLMs in complex systems, such as those used in telemedicine.
Comparing Fetal Liver Analysis with pure CNN vs. Complex Preprocessing Pipelines Felipe Soares Muylaert Barroso, Alex Sandro Roschildt Pinto, Aldo von Wangenheim, Luis Otavio Santos, Karine Souza da Correggio, et al. Proceedings IEEE Chilean Conference on Electrical Electronics Engineering Information and Communication Technologies Chilecon, 2025
The Internet of Flying Things Daniel Fernando Pigatto, Mariana Rodrigues, João Vitor de Carvalho Fontes, Alex Sandro Roschildt Pinto, James Smith, et al. Internet of Things A to Z Technologies and Applications, 2025
Training Swinging Door Trending Compression Algorithm for IoT devices Juan David Arias Correa, Alex Sandro Roschildt Pinto, Carlos Montez 2024 IEEE 10th World Forum on Internet of Things Wf Iot 2024, 2024 The communication system is essential for the devices used in IoT. Nevertheless, the consumption caused by transmitting and receiving data is a concern in many scenarios. That is due to these devices commonly having limited energy and computational capacities. Transmission of redundant or irrelevant samples frequently wastes the device’s resources. Furthermore, Storing a large amount of redundant data could consume storage space and not offer any benefit. Data Compression (DC) methods are a potential solution. DC could reduce communication usage and storage demand. This research proposes the Training Swing Door Trending (TSDT) for being implemented in IoT Devices. TSDT is a new algorithm that improves the classic Swing Door Trending (STD). They represent the data by trend lines and have a constant computational complexity. TSDT has a training step for the automatic configuration of its parameters. This article additionally presents the Compression Factor (C-Score), a new quality metric to analyze the compression results in lossy DC methods. C-Score takes as a basis the F-Score, a measure of predictive performance. The C-Score uses the Compression and Error metrics to evaluate the compression performance in Lossy Algorithms.
Practical Interface Diversity for Improving Network Performance in IEEE 802.15.4 Bruno Monteiro Pires, Carlos A. Astudillo, Alex Roschildt Pinto, Alexandro Baldassin 2024 IEEE Latin American Conference on Communications Latincom 2024 Proceedings, 2024 Wireless sensor networks, as well as other low power wireless systems for the Internet of Things (IoT), had always struggled to deal with the highly unstable radio frequency links over which they are built. Various approaches aiming at reducing the impact of low quality links on the lifespan and performance of IoT devices have already been proposed. However, only few of them provide practical improvements in data link performance, usually by detecting and avoiding temporarily bad links. This paper aims at exploring alternatives to mitigate link instability issues through the use of interface diversity. We propose PhyNode, a wireless sensor node architecture intended to complement generic models by adding support for interface diversity. Moreover, we introduce PhyMAC, a possible amendment to the IEEE 802.15.4 MAC sublayer to provide the capabilities needed to handle interface diversity. Results from experiments conducted with PhyNode prototypes evince that the number of interfaces supported by the current revision of the IEEE 802.15.4 standard is enough to allow the implementation of frequency-based interface diversity. The unmanaged use of multiple radio interfaces impose, however, great power consumption overheads. By using PhyMAC link quality estimation algorithms, dynamic interface selection and improved PhyNode’s energy consumption can be achieved. These results show that the use of the proposed algorithms in conjunction with interface diversity increases the robustness of the IoT communication links, with reduced power consumption overhead.
Classification using jumping emerging patterns and cosine similarity Proceedings of the 2015 International Conference on Artificial Intelligence Icai 2015 Worldcomp 2015, 2019
The Internet of Flying Things Daniel Fernando Pigatto, Mariana Rodrigues, João Vitor de Carvalho Fontes, Alex Sandro Roschildt Pinto, James Smith, et al. Internet of Things A to Z Technologies and Applications, 2018
RECENT SCHOLAR PUBLICATIONS
A Novel Cryptocurrency Trend Prediction Framework Powered by Innovative Feature Engineering R de Aquino Silva, ASR Pinto, GM De Araujo, LEM Ricci IEEE Access , 2025 2025
From Knowledge Extraction to Assertive Response: An LLM Chatbot for Information Retrieval in Telemedicine Systems BD Pupo, DG Costa, R Immich, A Wangenheim, ASR Pinto, ... Applied Sciences 15 (21), 11732 , 2025 2025 Citations: 1
Training Swinging Door Trending Compression Algorithm for IoT devices JDA Correa, ASR Pinto, C Montez 2024 IEEE 10th World Forum on Internet of Things (WF-IoT), 334-339 , 2024 2024 Citations: 2
MQTT-Chain: An MQTT approach using blockchain and smart contracts to achieve a new Quality of Service level BM Agostinho, S Chessa, R Perego, MAR Dantas, ASR Pinto 2024 11th International Conference on Future Internet of Things and Cloud … , 2024 2024 Citations: 3
An Architecture Proposal Using Hybrid Blockchain Applied for Supply Chain Tracking R Hoffmann, C Moratelli, ASR Pinto International Conference on Complex, Intelligent, and Software Intensive … , 2024 2024 Citations: 1
IoT Off-Grid, Data Collection from a Machine Learning Classification Using UAV A Goulart, ASR Pinto, A Boava, KRLJC Branco Sensors 22 (19), 7241 , 2022 2022 Citations: 2
Development of an efficiency platform based on MQTT for UAV controlling and DoS attack detection LM da Silva, HBB Menezes, MS Luccas, C Mailer, ASR Pinto, A Boava, ... Sensors 22 (17), 6567 , 2022 2022 Citations: 19
Lossy Data Compression for IoT Sensors: A Review JDA Correa, ASR Pinto, C Montez Internet of Things 19, 100516 , 2022 2022 Citations: 70
Data Collection in an IoT Off-Grid Environment Systematic Mapping of Literature A Goulart, ASR Pinto, A Boava, K Branco Sensors 22 (14), 5374 , 2022 2022 Citations: 2
Proposal of an Economy of Things Architecture and an Approach Comparing Cryptocurrencies BM Agostinho, MAR Dantas, ASR Pinto Sensors 21 (9), 3239 , 2021 2021 Citations: 1
An Approach Adopting Ethereum as a Wallet Domain Name System within the Economy of Things Context BM Agostinho, FB Pasini, FO Gomes, ASR Pinto, MAR Dantas 2020 IEEE/ACM 13th International Conference on Utility and Cloud Computing … , 2020 2020 Citations: 7
Iota vs. Ripple: A Comparison Inside An Economy of Things Architecture for Industry 4.0 BM Agostinho, MM Pereira, AP Back, ASR Pinto, MAR Dantas 2020 IEEE 6th World Forum on Internet of Things (WF-IoT), 1-6 , 2020 2020 Citations: 1
Omniconn: An architecture for heterogeneous devices interoperability on industrial Internet of Things BM Agostinho, CB de Souza, FO Gomes, ASR Pinto, MAR Dantas Advances on P2P, Parallel, Grid, Cloud and Internet Computing: Proceedings … , 2020 2020 Citations: 2
Swinging door trending compression algorithm for iot environments JDA Correa, ASR Pinto, C Montez, EM Leao Anais Estendidos do IX Simpósio Brasileiro de Engenharia de Sistemas … , 2019 2019 Citations: 19
The broadcast storm problem in FANETs and the Dynamic Neighborhood-based Algorithm as a countermeasure RDM Pires, ASR Pinto, KRLJC Branco IEEE Access 7, 59737-59757 , 2019 2019 Citations: 27
The Internet of Flying Things DF Pigatto, M Rodrigues, JV de Carvalho Fontes, ASR Pinto, J Smith, ... Internet of Things A to Z: Technologies and Applications, 529-562 , 2018 2018 Citations: 23
Data Summarization in the Node by Parameters (DSNP): Local Data Fusion in an IoT Environment L Maschi, A Pinto, R Meneguette, A Baldassin Sensors 18 (3), 799 , 2018 2018 Citations: 36
IoT Data Storage in the Cloud: A Case Study in Human Biometeorology B Vanelli, AR Pinto, MP da Silva, MAR Dantas, M Fazio, A Celesti, ... Cloud Infrastructures, Services, and IoT Systems for Smart Cities: Second … , 2018 2018 Citations: 2
Communication in Critical Embedded Systems: First Workshop, WoCCES 2013, Brasília, Brazil, May, 10, 2013, Second Workshop, WoCCES 2014, Florianópolis, Brazil, May 9, 2014 … K Branco, A Pinto, D Pigatto Springer , 2017 2017
UAV integration into IoIT: opportunities and challenges M Rodrigues, DF Pigatto, JV Fontes, AS Pinto, JP Diguet, KR Branco ICAS 2017 95 , 2017 2017 Citations: 23
MOST CITED SCHOLAR PUBLICATIONS
An approach to implement data fusion techniques in wireless sensor networks using genetic machine learning algorithms AR Pinto, C Montez, G Araújo, F Vasques, P Portugal Information Fusion 15, 90-101 , 2014 2014 Citations: 84
Lossy Data Compression for IoT Sensors: A Review JDA Correa, ASR Pinto, C Montez Internet of Things 19, 100516 , 2022 2022 Citations: 70
Avens-a novel flying ad hoc network simulator with automatic code generation for unmanned aircraft system EA Marconato, M Rodrigues, RM Pires, DF Pigatto, AR Pinto, KR Branco 2017 Citations: 58
IEEE 802.11 n vs. IEEE 802.15. 4: a study on Communication QoS to provide Safe FANETs EA Marconato, JA Maxa, DF Pigatto, ASR Pinto, N Larrieu, KRLJC Branco 2016 46th Annual IEEE/IFIP International Conference on Dependable Systems … , 2016 2016 Citations: 49
Internet of things data storage infrastructure in the cloud using NoSQL databases B Vanelli, MP da Silva, G Manerichi, ASR Pinto, MAR Dantas, ... IEEE Latin America Transactions 15 (4), 737-743 , 2017 2017 Citations: 37
Data Summarization in the Node by Parameters (DSNP): Local Data Fusion in an IoT Environment L Maschi, A Pinto, R Meneguette, A Baldassin Sensors 18 (3), 799 , 2018 2018 Citations: 36
HAMSTER-Healthy, mobility and security-based data communication architecture for Unmanned Aircraft Systems DF Pigatto, L Gonçalves, ASR Pinto, GF Roberto, JF Rodrigues Filho, ... 2014 International Conference on Unmanned Aircraft Systems (ICUAS), 52-63 , 2014 2014 Citations: 33
Outlier detection using k-means clustering and lightweight methods for Wireless Sensor Networks ATC Andrade, C Montez, R Moraes, AR Pinto, F Vasques, GL da Silva IECON 2016-42nd Annual Conference of the IEEE Industrial Electronics Society … , 2016 2016 Citations: 30
The broadcast storm problem in FANETs and the Dynamic Neighborhood-based Algorithm as a countermeasure RDM Pires, ASR Pinto, KRLJC Branco IEEE Access 7, 59737-59757 , 2019 2019 Citations: 27
Experimenting broadcast storm mitigation techniques in FANETs R de Melo Pires, SZ Arnosti, ASR Pinto, KRLJC Branco 2016 49th Hawaii International Conference on System Sciences (HICSS), 5868-5877 , 2016 2016 Citations: 26
Nodepm: a remote monitoring alert system for energy consumption using probabilistic techniques GPR Filho, J Ueyama, LA Villas, AR Pinto, VP Goncalves, G Pessin, ... Sensors 14 (1), 848-867 , 2014 2014 Citations: 26
Genetic machine learning approach for link quality prediction in mobile wireless sensor networks GM Araújo, AR Pinto, J Kaiser, LB Becker Cooperative robots and sensor networks, 1-18 , 2014 2014 Citations: 24
The Internet of Flying Things DF Pigatto, M Rodrigues, JV de Carvalho Fontes, ASR Pinto, J Smith, ... Internet of Things A to Z: Technologies and Applications, 529-562 , 2018 2018 Citations: 23
UAV integration into IoIT: opportunities and challenges M Rodrigues, DF Pigatto, JV Fontes, AS Pinto, JP Diguet, KR Branco ICAS 2017 95 , 2017 2017 Citations: 23
Design and implementation of a 6LoWPAN gateway for wireless sensor networks integration with the internet of things LF Schrickte, CB Montez, RSD Oliveira, ASR Pinto International Journal of Embedded Systems 8 (5-6), 380-390 , 2016 2016 Citations: 21
Development of an efficiency platform based on MQTT for UAV controlling and DoS attack detection LM da Silva, HBB Menezes, MS Luccas, C Mailer, ASR Pinto, A Boava, ... Sensors 22 (17), 6567 , 2022 2022 Citations: 19
Swinging door trending compression algorithm for iot environments JDA Correa, ASR Pinto, C Montez, EM Leao Anais Estendidos do IX Simpósio Brasileiro de Engenharia de Sistemas … , 2019 2019 Citations: 19
Integration of wireless sensor networks to the internet of things using a 6LoWPAN gateway LF Schrickte, C Montez, R De Oliveira, AR Pinto 2013 III Brazilian Symposium on Computing Systems Engineering, 119-124 , 2013 2013 Citations: 19
The HAMSTER Data Communication Architecture for Unmanned Aerial, Ground and Aquatic Systems: Aims, Scope and Definitions DF Pigatto, L Gonçalves, GF Roberto, JF Rodrigues Filho, ... Journal of Intelligent & Robotic Systems 84 (1-4), 705-723 , 2016 2016 Citations: 18
QK-means: a clustering technique based on community detection and K-means for deployment of cluster head nodes LN Ferreira, AR Pinto, L Zhao The 2012 International Joint Conference on Neural Networks (IJCNN), 1-7 , 2012 2012 Citations: 16
