A Deep Survey on Causal-Aware Temporal Graph Deep Learning Framework for Sepsis Detection

Read full text Summarise with VIRAT
Authors: Deepika K, Dr. Ananth Kumar T, Dr. Kanimozhi P
Cited by: 0
References: 25

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Due to its quick progression and high death rate, sepsis a potentially fatal clinical condition brought on by a irregular host response to infection remains a Substantial barrier in intensive care units. Sepsis must be detected early and accurately, but traditional machine learning models and rule-based scoring systems frequently fall short of capturing the intricate temporal dynamics, inter-variable dependencies, and causal mechanisms underlying the disease's progression. This deep survey focuses on Causal-Aware Temporal Graph Deep Learning frameworks for sepsis detection in order to overcome these constraints. In order to better represent time-evolving physiological signals and interactions among clinical variables, the survey examines recent developments that incorporate temporal modelling, graph neural networks, and causal inference. While graph-based learning captures the relationships between organ systems and biomarkers, temporal deep learning models efficiently learn the evolution of patient states. By differentiating between correlation and actual causal effects, the application of causal reasoning improves interpretability, robustness, and clinical trust. Furthermore, explainable AI approaches are explored to show how these frameworks facilitate transparent decision-making for medical professionals. The datasets, model architectures, learning techniques, and assessment metrics used in sepsis prediction are all methodically examined in this survey, which also identifies present issues and potential future research areas. All things considered, causal-aware temporal graph deep learning frameworks have a great deal of promise to enhance early sepsis detection, lower false alarms, and facilitate prompt clinical intervention all of which will eventually improve patient outcomes in critical care settings.

Fields of Study & Subjects

Sepsis Diagnosis and Treatment Machine Learning in Healthcare Artificial Intelligence in Healthcare

Publication details

Journal / source
Irish Interdisciplinary Journal of Science & Research
Volume
10
Issue
01
Pages
51–60
Publication date
1 January 2026
Language
EN