Artificial intelligence for oncology sepsis risk and mortality prediction
Rashid, Asrar Hafez, Wael Zaki, Syed Hashmi, Shahrukh
Rashid, Asrar
Hafez, Wael
Zaki, Syed
Hashmi, Shahrukh
Supervisor
Department
Computer Vision
Embargo End Date
Type
Book chapter
Date
License
Language
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Sepsis remains a leading cause of morbidity and mortality in oncology patients, occurring at higher rates due to immunosuppression from chemotherapy, immunotherapy, and stem cell transplantation. Traditional diagnostics lack sensitivity and real-time prediction, delaying treatment. Integrating artificial intelligence (AI) into sepsis prediction can transform risk stratification, early detection, and decision-making. This chapter explores AI-driven machine learning and deep learning models analyzing electronic health records, multiomics data, and physiological monitoring to enhance prediction. AI-powered risk models surpass conventional systems, enabling real-time surveillance, automated alerts, and precision medicine. Algorithms forecast sepsis onset, mortality risk, and complications, optimizing interventions, and antimicrobial stewardship. Challenges include model interpretability, data variability, and limited oncology-specific models. Addressing these with standardized data, multicenter validation, and explainable AI is crucial. Future AI-driven sepsis care will integrate multiomics, personalized pathways, and predictive analytics to improve survival and optimize resource allocation in immunocompromised oncology patients.
Citation
A. Rashid, W. Hafez, S. Zaki, S. Hashmi, "Artificial intelligence for oncology sepsis risk and mortality prediction," in Artificial Intelligence for Enhanced Diagnosis in Oncology, Elsevier, 2026, pp. 217-232.
Source
Artificial Intelligence for Enhanced Diagnosis in Oncology
Conference
Keywords
Subjects
Source
Publisher
Elsevier