Early detection of sepsis using machine learning algorithms
Year: 2024
Type: article
Abstract: In the intensive care unit (ICU), bedside surveillance data can appropriately predict the onset of sepsis, probably saving lives and lowering costs by permitting early intervention. Sepsis triggers a complicated immune reaction to pathogenic microbes, which frequently leads to septic shock and organ failure. Early detection is essential, but the excessive-pressure environment of emergency rooms can stress clinical personnel. Suggest a machine learning-based support vector machine (ML-SVM) technique to address this. The goal is to offer a reliable prediction of sepsis onset by studying ICU monitoring records to uncover subtle developments and early warning signs. This technology-driven approach complements their clinical judgment by aiding healthcare experts in making timely, knowledgeable selections. The ML-SVM machine automates the prediction of sepsis onset with a sensitivity of 91 % and a specificity of 93 %, supplying an accuracy of 95.2 %. This excessive- Overall Performance version offers improvements over present-day techniques, assisting scientific employees in making informed choices faster and decreasing the chance of sepsis-related problems. By improving early detection and optimizing resource allocation, the ML-SVM technique can significantly reduce affected person effects, keep lives, lessen healthcare prices, and alleviate the workload on healthcare experts in crucial care settings.
Source: Alexandria Engineering Journal
Authors: Rasha M. Abd El-Aziz, Alanazi Rayan
Institution: Jouf University
Language: English
Subfield: Epidemiology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $1,300
OpenAlex ID: W4403564147