Applying Machine Learning Sampling Techniques to Address Data Imbalance in a Chilean COVID-19 Symptoms and Comorbidities Dataset

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(3), P. 1132 - 1132

Published: Jan. 23, 2025

Reliably detecting COVID-19 is critical for diagnosis and disease control. However, imbalanced data in medical datasets pose significant challenges machine learning models, leading to bias poor generalization. The dataset obtained from the EPIVIGILA system Chilean Epidemiological Surveillance Process contains information on over 6,000,000 patients, but, like many current datasets, it suffers class imbalance. To address this issue, we applied various algorithms, both with without sampling methods, compared them using different classification diagnostic metrics such as precision, sensitivity, specificity, likelihood ratio positive, odds ratio. Our results showed that applying methods improved metric values contributed models better Effectively managing crucial reliable diagnosis. This study enhances understanding of how techniques can improve reliability contribute patient outcomes.

Language: Английский

---

Predicting the hydrogen storage capacity of alumina pillared interlayer clays using interpretable ensemble machine learning

International Journal of Hydrogen Energy, Journal Year: 2025, Volume and Issue: 120, P. 354 - 364

Published: March 28, 2025

Language: Английский

---

A novel NEMONET framework for enhanced RCC detection and staging in CT images

Deleted Journal, Journal Year: 2025, Volume and Issue: 28(1)

Published: Jan. 15, 2025

This study introduces NemoNet, a novel deep-learning framework designed for the automated detection and staging of Renal Cell Carcinoma (RCC) in 3D CT images. Leveraging comprehensive HubMAP RCC dataset, NemoNet integrates encoder-decoder architecture with advanced radiomic feature analysis to enhance tumour segmentation accuracy. The model employs multi-objective loss function balance precision prediction, outperforming traditional architectures like U-Net ResNet. Evaluation metrics, including Dice Coefficient, sensitivity, specificity, indicate superior performance, achieving an accuracy 92% score 0.88. While demonstrates robust results, challenges remain handling variability imaging quality full interpretability. findings suggest that offers significant advancements staging, potential applications personalized oncology treatment planning.

Language: Английский

---

Prediction model for type 2 diabetes mellitus and its association with mortality using machine learning in three independent cohorts from South Korea, Japan, and the UK: a model development and validation study

EClinicalMedicine, Journal Year: 2025, Volume and Issue: 80, P. 103069 - 103069

Published: Jan. 18, 2025

Language: Английский

---

Dynamic mortality prediction in critically Ill children during interhospital transports to PICUs using explainable AI

npj Digital Medicine, Journal Year: 2025, Volume and Issue: 8(1)

Published: Feb. 17, 2025

Critically ill children who require inter-hospital transfers to paediatric intensive care units are sicker than other admissions and have higher mortality rates. Current transport practice primarily relies on early clinical assessments within the initial hours of transport. Real-time risk during is lacking due absence data-driven assessment tools. Addressing this gap, our research introduces PROMPT (Patient-centred Outcome monitoring Mortality PredicTion), an explainable end-to-end machine learning pipeline forecast 30-day risks. The integrates continuous time-series vital signs medical records with episode-specific data provide real-time prediction. results demonstrated that PROMPT, both random forest logistic regression models achieved best performance AUROC 0.83 (95% CI: 0.79-0.86) 0.81 0.76-0.85), respectively. proposed model has proof-of-principle in predicting transported providing individual-level interpretability transports.

Language: Английский

---

Improving Symptom‐Based Medical Diagnosis Using Ensemble Learning Approaches

Systems Research and Behavioral Science, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 20, 2025

ABSTRACT Symptoms‐based health checkers are emerging as digital tools in modern healthcare offering patients the ability to self‐assess their status by inputting symptoms and receiving diagnostic suggestions. These systems rely on machine learning models accurately predict medical conditions based symptom data. In this study, we explore effectiveness of various algorithms with a particular focus ensemble methods improve accuracy reliability checkers. We evaluate multiple models—Decision Trees, Support Vector Machines (SVM), Logistic Regression, variations (Bagging, Stacking)—across three distinct datasets: ‘Reference Dataset,’ ‘Cough‐DDX Dataset’ ‘Cough‐DDX2 Dataset.’ Our results demonstrate that models, especially Bagging Decision Trees SVM, significantly outperform individual terms accuracy, precision, recall, F1 score. also tested clinical use cases achieved excellent highlighting real‐world applicability potential our approaches.

Language: Английский

---

Frobenius deep feature fusion architecture to detect diabetic retinopathy

Deleted Journal, Journal Year: 2025, Volume and Issue: 7(3)

Published: March 2, 2025

Language: Английский

---

An Interpretable Machine Learning Model to Predict Hospitalizations

Clinical eHealth, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

---

Sex-Specific Ensemble Models for Type 2 Diabetes Classification in the Mexican Population

Diabetes Metabolic Syndrome and Obesity, Journal Year: 2025, Volume and Issue: Volume 18, P. 1501 - 1525

Published: May 1, 2025

Type 2 diabetes (T2D) is considered a global pandemic by the World Health Organization (WHO), with growing prevalence, particularly in Mexico. Accurate early diagnosis remains challenge, especially when accounting for biological sex-based differences. This study aims to enhance classification of T2D Mexican population applying sex-specific ensemble models combined genetic algorithm-based feature selection. A dataset 1787 patients (895 females, 892 males) analyzed. Data are split sex, and selection performed using GALGO, tool. Classification including Random Forest, K-Nearest Neighbor, Support Vector Machine, Logistic Regression trained evaluated. Ensemble stacking constructed separately each sex improve performance. The male-specific model achieved 94% specificity 96% sensitivity, while female-specific reached 90% sensitivity. Both demonstrated strong overall proposed represent clinically valuable approach personalized diagnosis. By identifying predictive features, this work supports development precision medicine tools tailored population. contributes improving diagnostic supporting more equitable approaches clinical settings.

Language: Английский

---

COVID-19 from symptoms to prediction: A statistical and machine learning approach

Computers in Biology and Medicine, Journal Year: 2024, Volume and Issue: 182, P. 109211 - 109211

Published: Sept. 28, 2024

Language: Английский

---