Otimização do processo de vacinação por meio de Machine Learning e dispositivos IoT: monitoramento de doses aplicadas e recomendações baseadas em dados

Resumo

The vaccination campaign plays a crucial role in protecting public health, especially during periods of epidemiological outbreaks and pandemics. However, data collection and management during this process, in particular the control of vial quantities, doses and temperature, continue to face significant challenges. Some of the data recording activities, such as monitoring the quantity of doses and temperature, are carried out manually, which can result in delays in service, recording errors and lack of efficiency. Analytical reports that synthesize data from records by remote sensing devices are generally used for visualization and decision making. In this context, this work presents an approach to optimize the monitoring process through a method that uses a model generated by Machine Learning to suggest recommendations based on the daily target and weather data from the OpenWeather API. The data was simulated using Internet of Things (IoT) devices with LoRa communication and RFID technology. Seven models were trained using GridSearchCV to optimize the process of tuning the best hyperparameters and selecting the best classification model. The Gradient Boosting algorithm achieved the best performance with an accuracy of approximately 96.3% and Best Parameters result: 'learning_rate': 0.1, 'max_depth': 3, 'n_estimators': 200. The data was retrieved and presented in a dashboard of Power BI. The use of Machine Learning and remote sensing technology in data management during the vaccination process represents a significant advance for public health. Automating collection, generating information and forecasting demands are fundamental to ensuring a quick and accurate response in emergency situations, such as outbreaks of contagious diseases. Furthermore, the technology used can be adapted and expanded to other healthcare contexts, providing greater efficiency and effectiveness in various areas of medicine, such as organ transplantation, for example

Descrição

Palavras-chave

., ., .

Citação

SANTOS, Jacó Miranda dos. Otimização do processo de vacinação por meio de Machine Learning e dispositivos IoT: monitoramento de doses aplicadas e recomendações baseadas em dados. 2024. 73 f. Dissertação (Mestrado em Informática) - Universidade Federal do Amazonas, Manaus (AM), 2024.

Avaliação

Revisão

Suplementado Por

Referenciado Por

Licença Creative Commons

Exceto quando indicado de outra forma, a licença deste item é descrita como Acesso Aberto