PSO-XnB: a proposed model for predicting hospital stay of CAD patients DOI Creative Commons
Geetha Pratyusha Miriyala, Arun Kumar Sinha

Frontiers in Artificial Intelligence, Journal Year: 2024, Volume and Issue: 7

Published: May 3, 2024

Coronary artery disease poses a significant challenge in decision-making when predicting the length of stay for hospitalized patient. This study presents predictive model—a Particle Swarm Optimized-Enhanced NeuroBoost—that combines deep autoencoder with an eXtreme gradient boosting model optimized using particle swarm optimization. The uses fuzzy set rules to categorize into four distinct classes, followed by data preparation and preprocessing. In this study, dimensionality is reduced neural autoencoders. reconstructed obtained from autoencoders given as input model. Finally, tuned optimization obtain optimal hyperparameters. With proposed technique, achieved superior performance overall accuracy 98.8% compared traditional ensemble models past research works. also scored highest other metrics such precision, recall, particularly F1 scores all categories hospital stay. These validate suitability our medical healthcare applications.

Language: Английский

In-hospital mortality, readmission, and prolonged length of stay risk prediction leveraging historical electronic patient records DOI Creative Commons
Rajeev Bopche, Lise Tuset Gustad, Jan Egil Afset

et al.

JAMIA Open, Journal Year: 2024, Volume and Issue: 7(3)

Published: July 1, 2024

This study aimed to investigate the predictive capabilities of historical patient records predict adverse outcomes such as mortality, readmission, and prolonged length stay (PLOS).

Language: Английский

Citations

4
Late fused multi-modal neural network with votingclassifier for Parkinson’s Disease detection DOI Creative Commons
Abeer Aljohani

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: March 4, 2024

Abstract Parkinson’s Disease (PD) is categorized as a neurodegenerative progressivedisease caused by the destruction of cells in midbrain posterior.Detecting PD its early stages will help physicians alleviate complications disease. Artificial Intelligence (AI) considered groupof trained models that can be used for classification and regression. Differentmodalities such text, speech, picture, detecting PD.This research proposes multi-modal deep learning recognition technique forPD classification. To improve quality detection stages,the proposed method composed three main sections. These sections are:feature extracting, merging, classifying. As feature extractors combination Convolutional Neural Network (CNN) attention mechanisms isdeveloped. extract features from related motion signals ofCNN Long-Short Term Memory (LSTM) model used. Finally, Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM),Extreme Boot Classifier (XGB), voting classifier are to distinguishbetween healthy subjects. The experimental result indicates 99.95%accuracy, 99.99% precision, 99.98% sensitivity, 99.95% F1-score usingthe CNN with on handwritingand corresponding datasets. achieved results show proposedmethod extracting both handwriting pictures correlatedmotor symptoms followed fusing finally using asthe achieve perfect performance

Language: Английский

Citations

0
Inhospital Mortality, Readmission, and Prolonged Length of Stay Risk Prediction Leveraging Historical Electronic Health Records DOI Creative Commons
Rajeev Bopche, Lise Tuset Gustad, Jan Egil Afset

et al.

medRxiv (Cold Spring Harbor Laboratory), Journal Year: 2024, Volume and Issue: unknown

Published: April 16, 2024

Abstract Objective The aim of this study was to investigate predictive capabilities historical records patients maintained at hospitals towards predicting an impending adverse outcomes such as, mortality, readmission, and prolonged length stay (PLOS). Methods Leveraging a de-identified dataset from tertiary care university hospital, we developed eXplainable Artificial Intelligence (XAI) framework combining tree-based traditional ML models with interpretations, statistical analysis predictors PLOS. Results Our demonstrated exceptional performance notable Area Under the Receiver Operating Characteristic (AUROC) 0.9625 Precision-Recall Curve (AUPRC) 0.8575 for 30-day mortality discharge AUROC 0.9545 AUPRC 0.8419 admission. For readmission PLOS risk highest achieved were 0.8198 0.9797 repectively. machine learning (ML) consistently outperformed in all four prediction tasks. key age, derived temporal features, routine laboratory tests, diagnostic procedural codes. Conclusion underscores potential leveraging medical history enhanced analytics hospitals. We present accurate intuitive early warning that can be easily implemented current developing digital health platforms accurately predict outcomes.

Language: Английский

Citations

0
PSO-XnB: a proposed model for predicting hospital stay of CAD patients DOI Creative Commons
Geetha Pratyusha Miriyala, Arun Kumar Sinha

Frontiers in Artificial Intelligence, Journal Year: 2024, Volume and Issue: 7

Published: May 3, 2024

Coronary artery disease poses a significant challenge in decision-making when predicting the length of stay for hospitalized patient. This study presents predictive model—a Particle Swarm Optimized-Enhanced NeuroBoost—that combines deep autoencoder with an eXtreme gradient boosting model optimized using particle swarm optimization. The uses fuzzy set rules to categorize into four distinct classes, followed by data preparation and preprocessing. In this study, dimensionality is reduced neural autoencoders. reconstructed obtained from autoencoders given as input model. Finally, tuned optimization obtain optimal hyperparameters. With proposed technique, achieved superior performance overall accuracy 98.8% compared traditional ensemble models past research works. also scored highest other metrics such precision, recall, particularly F1 scores all categories hospital stay. These validate suitability our medical healthcare applications.

Language: Английский

Citations

0