Development of Algorithms for Predicting Clinical Outcomes in Patients with Traumatic Brain Injury using Non-Invasive Intracranial Pressure Monitoring and Machine Learning Related Parameters.

• Grant number: 23/14666-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2024
• End date: April 30, 2025
• Field of knowledge: Health Sciences - Medicine - Surgery
• Principal Investigator: Wellingson Silva Paiva
• Grantee: Lucca Biolcati Palavani
• Host Institution: Faculdade de Medicina (FM). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Traumatic brain injury (TBI) is a global public health concern, primarily affecting the active age group of the population. In both the United States and Brazil, TBI stands as one of the leading causes of morbidity and mortality, leaving thousands of patients with significant sequelae after the trauma. This study aims to enhance the prediction of neurological outcomes in TBI patients by employing non-invasive intracranial pressure (ICP) monitoring methods and advanced Machine Learning (ML) techniques. The project will gather data from non-invasive ICP monitoring, coupled with clinical information. ML algorithms, including linear and logistic regression, random forest, neural networks, SVM, decision trees, PCA, clustering algorithms, and recurrent neural networks, will be applied to develop precise prediction models. The objective is to establish an accurate and personalized system for forecasting clinical outcomes. This advancement will not only refine medical interventions but also improve treatment plans for individual patients.The algorithmic results will be meticulously compared with reference models based on invasive methods of ICP monitoring. Rigorous statistical analyses will be conducted to validate the accuracy of the developed models. This project marks an innovative approach to managing TBI patients, seamlessly integrating non-invasive ICP monitoring technologies with ML. The result is a comprehensive solution offering personalized medicine and significantly enhancing the clinical outcomes of patients.