Introduction: The early prediction of sepsis based on machine learning or deep learning has achieved good results.Most of the methods use structured data stored in electronic medical records, but the pathological characteristics of sepsis involve complex interactions between multiple physiological systems and signaling pathways, resulting in mixed structured data. Some researchers will introduce unstructured data when also introduce confounders. These confounders mask the direct causality of sepsis, leading the model to learn misleading correlations. Finally, it affects the generalization ability, robustness, and interpretability of the model.
Methods: To address this challenge, we propose an early sepsis prediction approach based on causal inference which can remove confounding effects and capture causal relationships. First, we analyze the relationship between each type of observation, confounder, and label to create a causal structure diagram. To eliminate the effects of different confounders separately, the methods of back-door adjustment and instrumental variable are used. Specifically, we learn the confounder and an instrumental variable based on mutual information from various observed data and eliminate the influence of the confounder by optimizing mutual information. We use back-door adjustment to eliminate the influence of confounders in clinical notes and static indicators on the true causal effect.
Results: Our method, named CISepsis, was validated on the MIMIC-IV dataset. Compared to existing state-of-the-art early sepsis prediction models such as XGBoost, LSTM, and MGP-AttTCN, our method demonstrated a significant improvement in AUC. Specifically, our model achieved AUC values of 0.921, 0.920, 0.919, 0.923, 0.924, 0.926, and 0.926 at the 6, 5, 4, 3, 2, 1, and 0 time points, respectively. Furthermore, the effectiveness of our method was confirmed through ablation experiments.
Discussion: Our method, based on causal inference, effectively removes the influence of confounding factors, significantly improving the predictive accuracy of the model. Compared to traditional methods, this adjustment allows for a more accurate capture of the true causal effects of sepsis, thereby enhancing the model's generalizability, robustness, and interpretability. Future research will explore the impact of specific indicators or treatment interventions on sepsis using counterfactual adjustments in causal inference, as well as investigate the potential clinical application of our method.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11638194 | PMC |
http://dx.doi.org/10.3389/fcimb.2024.1488130 | DOI Listing |
Ophthalmol Glaucoma
January 2025
Eye Institute and Department of Ophthalmology, Eye & ENT Hospital, Fudan University, Shanghai, 200031, China; NHC Key Laboratory of Myopia and Related Eye Diseases; Key Laboratory of Myopia and Related Eye Diseases, Chinese Academy of Medical Sciences, Shanghai, 200031, China; Shanghai Key Laboratory of Visual Impairment and Restoration, Shanghai, 200031, China; State Key Laboratory of Medical Neurobiology, Fudan University, Shanghai, 200032, China. Electronic address:
Purpose: Liver disease is associated with a range of extrahepatic complications, which have recently been expanded to include ophthalmic conditions. However, evidence is lacking regarding its impact on primary open-angle glaucoma (POAG). This study aimed to investigate whether major liver diseases, including metabolic dysfunction-associated steatotic liver disease (MASLD), alcoholic liver disease (ALD), viral hepatitis, and liver fibrosis and cirrhosis, were associated with POAG.
View Article and Find Full Text PDFJ Trace Elem Med Biol
December 2024
Department of Biochemistry, Christian Medical College, Vellore, Tamil Nadu, India; Affiliated to The Tamil Nadu Dr. MGR Medical University, Chennai, India. Electronic address:
Introduction: Observational studies have found that higher iron levels are associated with an increased risk of diabetes mellitus. Given the limitations of causal inferences from observational studies and the expensive and time-consuming nature of randomized controlled trials, Mendelian randomization analysis presents a reasonable alternative to study causal relationships. Previous MR analyses studying iron levels and diabetes have used indirect markers of iron levels, such as serum ferritin, and found conflicting results.
View Article and Find Full Text PDFJ Clin Anesth
January 2025
Department of Anesthesia, Critical Care and Pain Medicine, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114, USA. Electronic address:
Study Objective: To assess whether, in a lung resection cohort with a low probability of confounding by indication, higher FiO is associated with an increased risk of impaired postoperative oxygenation - a clinical manifestation of lung injury/dysfunction.
Design: Pre-specified registry-based retrospective cohort study.
Setting: Two large academic hospitals in the United States.
BMJ Open
January 2025
Faculty of Medical Sciences, University College London, London, UK.
Introduction: An ageing population and a workforce crisis have triggered an ambitious UK strategy for sustained delivery of healthcare. In perioperative care (the management of patients from contemplation of surgery until full recovery), it is recognised that interventions are needed to place the workforce on a more sustainable footing through cross-functionality and skill-shifting, namely with advanced practice roles. However, despite some reports and reviews in the literature, it is unclear how skills development efforts may potentially support workforce transformation for an effective and resilient perioperative care workforce.
View Article and Find Full Text PDFEBioMedicine
January 2025
Novo Nordisk Foundation Center for Basic Metabolic Research, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark. Electronic address:
Background: Lipid species are emerging as biomarkers for cardiometabolic risk in both adults and children. The genetic regulation of lipid species and their impact on cardiometabolic risk during early life remain unexplored.
Methods: Using mass spectrometry-based lipidomics, we measured 227 plasma lipid species in 1149 children and adolescents (44.
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!