Optimizing Sepsis Care Through CNN-LSTM Models: A Comprehensive Data-Driven Approach to Enhance Early Detection and Management

DOI

10.2991/978-94-6463-700-7_16How to use a DOI?

Keywords

Sepsis; Neonatal Sepsis; Machine Learning; Artificial Intelligence; Early Diagnosis; CNN-LSTM; Predictive Modelling; Clinical Decision Support; Healthcare Innovation

Abstract

Sepsis continues to be a global threat affecting mainly neonate and immunocompromised with high complications and mortality rates. Even with the availability of increasingly sophisticated diagnostic tests and therapeutic approaches, the current ability to identify patients in whom sepsis will develop remains low, partly attributable to the fact that this condition is polymorphic and progresses quickly. Taking part in this research is the application of machine learning (ML) and artificial intelligence (AI) to improve sepsis early diagnosis and management. The research also focuses on outcome estimation by comparing the advanced models, specifically the CNN-LSTM architecture based on time series clinic variables, vital signs, and laboratory results. The CNN-LSTM model showed higher accurate, precise, recall and AUC-ROC than other models in all datasets to provide early recognition of sepsis onset before the clinical sign is more prominent. How clinical pathways need to be interoperated with AI solutions with an aim of supporting interventions whenever necessary and enhance client experiences. Further, this research explores crucial issues: data preprocessing, model explain ability, and clinician endorsement with the focus on the methods, quality of training, and essential ethical aspects that should be considered. Here, the paper applies the strengths of AI—the capacity to identify nuanced patterns and update one’s mode of thinking—to advance the research field of predictive health care. Building on these insights, subsequent advancements focus on the AI and DL in sepsis diagnosis, and treatment option with an inclusion on adverse effects vulnerable groups.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Vaibhav C. Gandhi
AU  - Jaina Patel
AU  - Bhumika Prajapati
AU  - Frenisha Jaimish Digaswala
AU  - Roma Vishal Barot
AU  - Nirav Patel
PY  - 2025
DA  - 2025/04/19
TI  - Optimizing Sepsis Care Through CNN-LSTM Models: A Comprehensive Data-Driven Approach to Enhance Early Detection and Management
BT  - Proceedings of the International Conference on Advancements in Computing Technologies and Artificial Intelligence (COMPUTATIA-2025)
PB  - Atlantis Press
SP  - 187
EP  - 204
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-700-7_16
DO  - 10.2991/978-94-6463-700-7_16
ID  - Gandhi2025
ER  -