Cloud-based HPC applied to the scalability of a patient clinical deterioration risk prediction system via machine learning

• Grant number: 23/00744-1
• Support Opportunities: Regular Research Grants
• Start date: April 01, 2024
• End date: March 31, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Agreement: MCTI/MC
• Principal Investigator: Rhanna Emanuela Fontenele Lima de Carvalho
• Grantee: Rhanna Emanuela Fontenele Lima de Carvalho
• Host Institution: Centro de Ciências da Saúde (CCS). Universidade Estadual do Ceará (UECE). Campus do Itaperi. Fortaleza , SP, Brazil
• Associated researchers: Francisco Heron de Carvalho Junor ; Melissa Soares Medeiros ; Ticiana Linhares Coelho da Silva ; Victor Almeida Campos
• Associated scholarship(s): 25/12449-0 - Cloud-based HPC Applied to the Scalability of a Patient Clinical Deterioration Risk Prediction System via Machine Learning, BP.TT; 25/02067-2 - Machine Learning-Based Clinical Patient Deterioration Risk Prediction Model, BP.TT

Abstract

The purpose of this research is to propose a risk-alert model for patients undergoing clinical deterioration using machine learning, based on data recorded by health professionals in electronic medical records, forms completed on paper by nurses, laboratory tests, and demographic information. Technological development research, carried out in eight stages, ranging from the collection of information to the generation of the risk prediction model. The proposed risk model will be trained based on health indicators collected from medical records and may have the following categories: red (high risk), yellow (moderate risk), and green (low risk). The model can also be a regression model that allows the prediction of a numerical score to be later associated with one of these risk ranges. This research is expected to provide valid, reliable, and scalable technology that can predict the clinical deterioration of patients in a hospital network within 72 h, and thus assist the health team in making early decisions. In addition, it is expected to strengthen collaboration between national and international research groups, including computer science and health science researchers, and the government through the Department of Health of the State of Ceará, in the search for a quality public health care system. (AU)