Version 1
: Received: 15 April 2023 / Approved: 18 April 2023 / Online: 18 April 2023 (05:20:36 CEST)
How to cite:
Cruz Castañeda, W.A.; Bertemes Filho, P. Towards an Artificial Intelligence Based Chronic Disease Management. Preprints2023, 2023040491. https://doi.org/10.20944/preprints202304.0491.v1
Cruz Castañeda, W.A.; Bertemes Filho, P. Towards an Artificial Intelligence Based Chronic Disease Management. Preprints 2023, 2023040491. https://doi.org/10.20944/preprints202304.0491.v1
Cruz Castañeda, W.A.; Bertemes Filho, P. Towards an Artificial Intelligence Based Chronic Disease Management. Preprints2023, 2023040491. https://doi.org/10.20944/preprints202304.0491.v1
APA Style
Cruz Castañeda, W.A., & Bertemes Filho, P. (2023). Towards an Artificial Intelligence Based Chronic Disease Management. Preprints. https://doi.org/10.20944/preprints202304.0491.v1
Chicago/Turabian Style
Cruz Castañeda, W.A. and Pedro Bertemes Filho. 2023 "Towards an Artificial Intelligence Based Chronic Disease Management" Preprints. https://doi.org/10.20944/preprints202304.0491.v1
Abstract
Chronic non-communicable diseases (NCDs) are major public health problems and a significant financial burden on public health systems. By 2030, mortality due to NCDs, such as cardiovascular diseases, cancer, respiratory diseases, and diabetes, is predicted to increase in Brazil. A peculiar aspect of NCDs involves their long-term and integrated care management. This paper proposes a chronic disease management platform based on artificial intelligence to deliver digital health services everywhere. The proposed platform is anchored and built with healthcare 4.0 technologies, such as wearable devices, the internet of medical things, and artificial intelligence cloud-based solutions that allow the deployment of a smart healthcare system. In addition, the paper presents the feasibility of the platform in a diabetes prediction study case. For the study case, an initial dataset was established with bio-impedance, oxygen concentration, pulse rate, skin impedance, and skin temperature attributes. A baseline was implemented with ten regression models to assess the prediction performance of the mean squared error, root mean squared error, and r-squared score to compare predictive findings with capillary blood glucose measurements. Results evidence that the decision tree regressor and three ensemble methods (bagging decision tree regressor, random forest regressor, and AdaBoost regressor) yielded improvement over the other models. Moreover, a comparison among those models revealed that the decision tree regressor outperforms them and presents promissory outcomes.
Keywords
Smart Health; Internet of Medical Things; Healthcare 4.0; Chronic-Disease Management; Artificial Intelligence
Subject
Engineering, Electrical and Electronic Engineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.