The study aims to develop AICare, an interpretable mortality prediction model, using electronic medical records (EMR) from follow-up visits for end-stage renal disease (ESRD) patients. AICare includes a multichannel feature extraction module and an adaptive feature importance recalibration module. It integrates dynamic records and static features to perform personalized health context representation learning. The dataset encompasses 13,091 visits and demographic data of 656 peritoneal dialysis (PD) patients spanning 12 years. An additional public dataset of 4,789 visits from 1,363 hemodialysis (HD) patients is also considered. AICare outperforms traditional deep learning models in mortality prediction while retaining interpretability. It uncovers mortality-feature relationships and variations in feature importance and provides reference values. An AI-doctor interaction system is developed for visualizing patients' health trajectories and risk indicators.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10724364 | PMC |
| http://dx.doi.org/10.1016/j.patter.2023.100892 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Informing ICU Digital Twins: Dynamic Assessment of Cardiorespiratory Failure Trajectories in Patients with Sepsis.
Shock
January 2025
Department of Industrial and Systems Engineering, University of Florida, P.O. Box 116595, Gainesville, FL, 32611, USA.
Understanding clinical trajectories of sepsis patients is crucial for prognostication, resource planning, and to inform digital twin models of critical illness. This study aims to identify common clinical trajectories based on dynamic assessment of cardiorespiratory support using a validated electronic health record data that covers retrospective cohort of 19,177 patients with sepsis admitted to ICUs of Mayo Clinic Hospitals over eight-year period. Patient trajectories were modeled from ICU admission up to 14 days using an unsupervised machine learning two-stage clustering method based on cardiorespiratory support in ICU and hospital discharge status.
View Article and Find Full Text PDFDysregulated Clot Mechanics and Kinetics Impacted by Injury Severity, Predict Mortality After Trauma.
Shock
January 2025
Department of Biomedical Engineering, Rutgers University, Piscataway, NJ 599 Taylor Road, Room 209, Piscataway, NJ, USA 08854.
Introduction: Coagulopathy following traumatic injury impairs stable blood clot formation and exacerbates mortality from hemorrhage. Understanding how these alterations impact blood clot stability is critical to improving resuscitation. Furthermore, the incorporation of machine learning algorithms to assess clinical markers, coagulation assays and biochemical assays allows us to define the contributions of these factors to mortality.
View Article and Find Full Text PDFMachine Learning Potential for Identifying and Forecasting Complex Environmental Drivers of Infections in the United States.
Environ Health Perspect
January 2025
Centre for Environment, Fisheries and Aquaculture Science (CEFAS), Weymouth, UK.
Background: Environmental change in coastal areas can drive marine bacteria and resulting infections, such as those caused by , with both foodborne and nonfoodborne exposure routes and high mortality. Although ecological drivers of in the environment have been well-characterized, fewer models have been able to apply this to human infection risk due to limited surveillance.
Objectives: The Cholera and Other Illness Surveillance (COVIS) system database has reported infections in the United States since 1988, offering a unique opportunity to both explore the forecasting capabilities machine learning could provide and to characterize complex environmental drivers of infections.
Genetic ancestry superpopulations show distinct prevalence and outcomes across pediatric central nervous system tumors from the PBTA and PNOC.
Neuro Oncol
January 2025
Division of Oncology, Children's Hospital of Philadelphia, Philadelphia, PA, USA.
Background: Central nervous system (CNS) tumors lead to cancer-related mortality in children. Genetic ancestry-associated cancer prevalence and outcomes have been studied, but is limited.
Methods: We performed genetic ancestry prediction in 1,452 pediatric patients with paired normal and tumor whole genome sequencing from the Open Pediatric Cancer (OpenPedCan) project to evaluate the influence of reported race and ethnicity and ancestry-based genetic superpopulations on tumor histology, molecular subtype, survival, and treatment.
A Novel Artificial Intelligence-Enhanced Digital Network for Prehospital Emergency Support: Community Intervention Study.
J Med Internet Res
January 2025
Department of Cardiology, Yonsei University College of Medicine, Seoul, Republic of Korea.
Background: Efficient emergency patient transport systems, which are crucial for delivering timely medical care to individuals in critical situations, face certain challenges. To address this, CONNECT-AI (CONnected Network for EMS Comprehensive Technical-Support using Artificial Intelligence), a novel digital platform, was introduced. This artificial intelligence (AI)-based network provides comprehensive technical support for the real-time sharing of medical information at the prehospital stage.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!