Explainable Boosting Machines Identify Key Metabolomic Biomarkers in Rheumatoid Arthritis DOI Creative Commons
Fatma Hilal Yağın, Cemil Çolak, Abdulmohsen Algarni

и другие.

Medicina, Год журнала: 2025, Номер 61(5), С. 833 - 833

Опубликована: Апрель 30, 2025

Background and Objectives: Rheumatoid arthritis (RA) is a chronic autoimmune disease characterised by joint inflammation pain. Metabolomics approaches, which are high-throughput profiling of small molecule metabolites in plasma or serum RA patients, have so far provided biomarker discovery the literature for clinical subgroups, risk factors, predictors treatment response using classical statistical approaches machine learning models. Despite these recent developments, an explainable artificial intelligence (XAI)-based methodology has not been used to identify metabolomic biomarkers distinguish patients with RA. This study constructed XAI-based EBM model global metabolomics predictive develop classification that can from healthy controls. Materials Methods: Global data were analysed (49 samples) individuals (10 samples). SMOTE technique was class imbalance preprocessing. EBM, LightGBM, AdaBoost algorithms applied generate discriminatory between Comprehensive performance metrics calculated, interpretability optimal assessed local feature descriptions. Results: A total 59 samples analysed, 49 10 subjects. The generated better results than LightGBM attaining AUC 0.901 (95% CI: 0.847–0.955) 87.8% sensitivity helps prevent false negative early diagnosis. primary EBM-based XAI identified N-acetyleucine, pyruvic acid, glycerol-3-phosphate. explanation analysis indicated elevated acid levels significantly correlated RA, whereas N-acetyleucine exhibited nonlinear relationship, implying possible protective effects at specific concentrations. Conclusions: underscores promise evidence-based medicine developing through metabolomics. discovered offer significant insights into pathophysiology may function as diagnostic therapeutic targets. Incorporating methodologies integrated improves transparency increases applicability models diagnosis/management. Furthermore, transparent structure empowers clinicians understand verify reasoning behind each prediction, thereby fostering trust AI-assisted decision-making facilitating integration routine practice.

Язык: Английский

Boosting-Based Machine Learning Applications in Polymer Science: A Review DOI Open Access
Ivan Malashin, В С Тынченко, Andrei Gantimurov

и другие.

Polymers, Год журнала: 2025, Номер 17(4), С. 499 - 499

Опубликована: Фев. 14, 2025

The increasing complexity of polymer systems in both experimental and computational studies has led to an expanding interest machine learning (ML) methods aid data analysis, material design, predictive modeling. Among the various ML approaches, boosting methods, including AdaBoost, Gradient Boosting, XGBoost, CatBoost LightGBM, have emerged as powerful tools for tackling high-dimensional complex problems science. This paper provides overview applications science, highlighting their contributions areas such structure-property relationships, synthesis, performance prediction, characterization. By examining recent case on techniques this review aims highlight potential advancing characterization, optimization materials.

Язык: Английский

Процитировано

2
Machine Learning-Driven Optimization for Predicting Compressive Strength in Fly Ash Geopolymer Concrete DOI Creative Commons

Maryam Bypour,

Mohammad Yekrangnia, Mahdi Kioumarsi

и другие.

Cleaner Engineering and Technology, Год журнала: 2025, Номер unknown, С. 100899 - 100899

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

1
A novel machine learning-based approach to determine the reduction factor for punching shear strength capacity of voided concrete slabs DOI Creative Commons

Alireza Mahmoudian,

Mussa Mahmoudi, Mohammad Yekrangnia

и другие.

Deleted Journal, Год журнала: 2025, Номер 2(1)

Опубликована: Фев. 27, 2025

Язык: Английский

Процитировано

1
Interpretable machine learning models for predicting flexural bond strength between FRP/steel bars and concrete DOI Creative Commons
Mohsen Ebrahimzadeh,

Alireza Mahmoudian,

Nima Tajik

и другие.

Structures, Год журнала: 2025, Номер 74, С. 108587 - 108587

Опубликована: Март 4, 2025

Язык: Английский

Процитировано

0
Explainable Boosting Machines Identify Key Metabolomic Biomarkers in Rheumatoid Arthritis DOI Creative Commons
Fatma Hilal Yağın, Cemil Çolak, Abdulmohsen Algarni

и другие.

Medicina, Год журнала: 2025, Номер 61(5), С. 833 - 833

Опубликована: Апрель 30, 2025

Background and Objectives: Rheumatoid arthritis (RA) is a chronic autoimmune disease characterised by joint inflammation pain. Metabolomics approaches, which are high-throughput profiling of small molecule metabolites in plasma or serum RA patients, have so far provided biomarker discovery the literature for clinical subgroups, risk factors, predictors treatment response using classical statistical approaches machine learning models. Despite these recent developments, an explainable artificial intelligence (XAI)-based methodology has not been used to identify metabolomic biomarkers distinguish patients with RA. This study constructed XAI-based EBM model global metabolomics predictive develop classification that can from healthy controls. Materials Methods: Global data were analysed (49 samples) individuals (10 samples). SMOTE technique was class imbalance preprocessing. EBM, LightGBM, AdaBoost algorithms applied generate discriminatory between Comprehensive performance metrics calculated, interpretability optimal assessed local feature descriptions. Results: A total 59 samples analysed, 49 10 subjects. The generated better results than LightGBM attaining AUC 0.901 (95% CI: 0.847–0.955) 87.8% sensitivity helps prevent false negative early diagnosis. primary EBM-based XAI identified N-acetyleucine, pyruvic acid, glycerol-3-phosphate. explanation analysis indicated elevated acid levels significantly correlated RA, whereas N-acetyleucine exhibited nonlinear relationship, implying possible protective effects at specific concentrations. Conclusions: underscores promise evidence-based medicine developing through metabolomics. discovered offer significant insights into pathophysiology may function as diagnostic therapeutic targets. Incorporating methodologies integrated improves transparency increases applicability models diagnosis/management. Furthermore, transparent structure empowers clinicians understand verify reasoning behind each prediction, thereby fostering trust AI-assisted decision-making facilitating integration routine practice.

Язык: Английский

Процитировано

0