DSpace Arşivi :: by Yazar "Yagin, Fatma Hilal" değerine göre listeleniyor

Yazar "Yagin, Fatma Hilal" seçeneğine göre listele

Listeleniyor 1 - 20 / 33

A manta ray-bayesian optimization approach for hyperparameter-tuned convolutional neural networks in lung cancer classification
(Nature Portfolio, 2026) Samal, Sonali; Sunder, Shyam; Gadekellu, Thippa Reddy; Yagin, Fatma Hilal; El Shawi, Radwa; Ahmed, Nada
Lung cancer remains a global health challenge that is unavoidable. Despite the advances in lung cancer classification using deep learning models, the performance remains highly dependent on hyperparameter selection, whereas conventional grid or random search methods are often computationally inefficient in high-dimensional spaces. So, to address the issue, this paper presents a Convolutional Neural Network(CNN) which is hybridized by dual stage hyperparameter optimization techniques for lung cancer image classification. The approach integrates Bayesian Optimization (BO) and Manta Ray Foraging Optimization (MRFO) to efficiently explore and fine-tune a defined hyperparameter search space, including convolution filter count, learning rate, dense layer neurons, and dropout rate. Initially, Bayesian Optimization explores the search space by modeling the objective function with a Gaussian Process and selecting candidate hyperparameters via the Expected Improvement criterion. The best solution obtained is then further enhanced using MRFO, which incrementally refines the parameters through its chain, cyclone, and somersault foraging mechanisms. This two-step process strikes a balance between exploring the world and taking advantage of local resources. The CNN trained with the optimized hyperparameters achieved good accuracy in lung cancer image classification, demonstrating the potency of combining probabilistic modeling with bio-inspired optimization. Experimental results show that the proposed hybrid CNN method has a testing accuracy of 98%, which is better than that of many cutting-edge models. The results show that metaheuristic-based optimization could be useful in deep learning applications, especially in medical image analysis.
A Thirty-Minute Nap Enhances Performance in Running-Based Anaerobic Sprint Tests during and after Ramadan Observance
(Mdpi, 2022) Yagin, Fatma Hilal; Eken, Ozgur; Bayer, Ramazan; Salcman, Vaclav; Gabrys, Tomasz; Koc, Hurmuz; Eken, Ismihan
The purpose of this study was to determine the impact of a 30 min nap (N30) on the Running-Based Anaerobic Sprint Test (RAST) both during and after Ramadan. Ten physically active kickboxers (age: 21.20 +/- 1.61 years, height: 174.80 +/- 4.34 cm, body mass: 73.30 +/- 7.10 kg and body mass index (BMI): 24.00 +/- 2.21 kg/m(2)) voluntarily performed the RAST test after an N30 and in a no-nap condition (NN) during two experimental periods: the last ten days of Ramadan (DR) and similar to 3 weeks after Ramadan (AR). During each DR-NN, DR-N30, AR-NN and AR-N30 protocol, kickboxers performed RAST performance. A statistically significant difference was found between Ramadan periods (DR vs. AR) in terms of max power (W) (F = 80.93; p(1) < 0.001; eta(2)(p) = 0.89), minimum power (W) (F = 49.05; p(1) < 0.001; eta(2)(p) = 0.84), average power (W) (F = 83.79; p(1) < 0.001; eta(2)(p) = 0.90) and fatigue index (%) results (F = 11.25; p(1) = 0.008; eta(2)(p) = 0.55). In addition, the nap factor was statistically significant in terms of the max power (W) (F = 81.89; p(2) < 0.001; eta(2)(p) = 0.90), minimum power (W) (F = 80.37; p(2) < 0.001; eta(2)(p) = 0.89), average power (W) (F = 108.41; p(2) < 0.001; eta(2)(p) = 0.92) and fatigue index (%) results (F = 16.14; p(2) = 0.003; eta(2)(p) = 0.64). Taking a daytime nap benefits subsequent performance in RAST. The benefits of napping were greater after an N30 opportunity for DR and AR.
Accuracy is not enough: explainable boosting machine model and identification of candidate biomarkers for real-time sepsis risk assessment in the emergency department
(Bmc, 2025) Yagin, Fatma Hilal; Aygun, Umran; Colak, Cemil; Alkhalifa, Amal K.; Alzakari, Sarah A.; Aghaei, Mohammadreza
BackgroundSepsis poses a significant threat in emergency settings, necessitating tools for early and interpretable risk assessment. This study aimed to develop a robust explainable boosting machine (EBM) model, one of the explainable artificial intelligence (XAI) technologies, to construct a predictive model that balances high accuracy and clinical interpretability for use in emergency departments (EDs) and to examine candidate biomarkers.MethodsThe study identified a significant class imbalance problem in the sepsis distribution among 560 sepsis and 1012 non-sepsis patients. To address the imbalance issue, SMOTE-NC was applied in the training data. The data was divided into two parts, 80% training and 20% testing. To ensure the reliability of the models and to report unbiased results, this process was repeated 100 times and the average performance was reported. To determine the best model for sepsis prediction, five different models (AdaBoost, Gradient Boosting, CatBoost, LightGBM, and EBM) were trained, and their performances were evaluated. In the last stage, we presented local and global explanations of EBM.ResultsThe EBM model achieved the highest success by reaching 79.1% F1-score, 80.9% sensitivity, and 84.8% AUC after resampling. In the global explanations, the variables with the highest weights in the model's decision process were identified as positive blood culture, oxygen saturation, and procalcitonin, respectively.ConclusionThe EBM model accurately predicts sepsis risk based on clinically relevant biomarkers. The model's high performance and inherent transparency can foster trust among clinicians and facilitate its integration into emergency department workflows for real-time decision support.
Assessment of Sepsis Risk at Admission to the Emergency Department: Clinical Interpretable Prediction Model
(Mdpi, 2024) Aygun, Umran; Yagin, Fatma Hilal; Yagin, Burak; Yasar, Seyma; Colak, Cemil; Ozkan, Ahmet Selim; Ardigo, Luca Paolo
This study aims to develop an interpretable prediction model based on explainable artificial intelligence to predict bacterial sepsis and discover important biomarkers. A total of 1572 adult patients, 560 of whom were sepsis positive and 1012 of whom were negative, who were admitted to the emergency department with suspicion of sepsis, were examined. We investigated the performance characteristics of sepsis biomarkers alone and in combination for confirmed sepsis diagnosis using Sepsis-3 criteria. Three different tree-based algorithms-Extreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Adaptive Boosting (AdaBoost)-were used for sepsis prediction, and after examining comprehensive performance metrics, descriptions of the optimal model were obtained with the SHAP method. The XGBoost model achieved accuracy of 0.898 (0.868-0.929) and area under the ROC curve (AUC) of 0.940 (0.898-0.980) with a 95% confidence interval. The five biomarkers for predicting sepsis were age, respiratory rate, oxygen saturation, procalcitonin, and positive blood culture. SHAP results revealed that older age, higher respiratory rate, procalcitonin, neutrophil-lymphocyte count ratio, C-reactive protein, plaque, leukocyte particle concentration, as well as lower oxygen saturation, systolic blood pressure, and hemoglobin levels increased the risk of sepsis. As a result, the Explainable Artificial Intelligence (XAI)-based prediction model can guide clinicians in the early diagnosis and treatment of sepsis, providing more effective sepsis management and potentially reducing mortality rates and medical costs.
Can Environmental Enrichment Mitigate Cell Apoptosis in the Central Nervous System Under Adverse Health Conditions? A Systematic Review
(Wiley, 2026) de Sousa Fernandes, Matheus Santos; Ferreira, Diorginis Jose Soares; Yagin, Fatma Hilal; Badicu, Georgian; Santos, Gabriela Carvalho Jurema; Guedes, Maria Carolina Santos; Ardigo, Luca Paolo
To summarize the available evidence in the literature regarding the effects of exposure to an enriched environment (EE) on the modulation of apoptotic markers in central tissue of rodents under unfavorable conditions. Searches were conducted in three databases: PubMed/Medline (70 articles), Scopus (65 articles), and EMBASE (128 articles), all of which were subjected to eligibility criteria. Of the 263 articles found, 95 duplicates were removed. After evaluating the title and abstract, 147 studies were excluded, leaving 21 articles, 16 of which were included in this systematic review. EE was implemented using various inanimate objects. The disruptive event/condition included social isolation, cerebral ischemia, stroke, postpartum depression, accelerated aging, ischemia/reperfusion, high altitude, Alzheimer's disease, hypobaric hypoxia, infrasound exposure, sepsis, and exposure to sevoflurane. Regarding the expression of apoptotic markers, after EE exposure, there was a reduction in the expression of p-IKK beta/IKK beta and p-P65/P65 in the hippocampus, a reduction in Bax, as well as cleaved Caspase-3, Cytochrome C, a reduction in the Bax/Bcl-2 ratio, and p53. As for anti-apoptotic markers, increased expression of Bcl-2 was observed following EE exposure. This systematic review concludes that the benefits of EE can reduce apoptosis through the modulation of both pro-apoptotic and anti-apoptotic genes. In this regard, EE has been shown to be an important neuroprotective tool in various adverse conditions, mitigating cognitive deficits by reducing apoptosis related to cellular stress.
Cancer Metastasis Prediction and Genomic Biomarker Identification through Machine Learning and eXplainable Artificial Intelligence in Breast Cancer Research
(Mdpi, 2023) Yagin, Burak; Yagin, Fatma Hilal; Colak, Cemil; Inceoglu, Feyza; Kadry, Seifedine; Kim, Jungeun
Aim: Method: This research presents a model combining machine learning (ML) techniques and eXplainable artificial intelligence (XAI) to predict breast cancer (BC) metastasis and reveal important genomic biomarkers in metastasis patients. Method: A total of 98 primary BC samples was analyzed, comprising 34 samples from patients who developed distant metastases within a 5-year follow-up period and 44 samples from patients who remained disease-free for at least 5 years after diagnosis. Genomic data were then subjected to biostatistical analysis, followed by the application of the elastic net feature selection method. This technique identified a restricted number of genomic biomarkers associated with BC metastasis. A light gradient boosting machine (LightGBM), categorical boosting (CatBoost), Extreme Gradient Boosting (XGBoost), Gradient Boosting Trees (GBT), and Ada boosting (AdaBoost) algorithms were utilized for prediction. To assess the models' predictive abilities, the accuracy, F1 score, precision, recall, area under the ROC curve (AUC), and Brier score were calculated as performance evaluation metrics. To promote interpretability and overcome the black box problem of ML models, a SHapley Additive exPlanations (SHAP) method was employed. Results: The LightGBM model outperformed other models, yielding remarkable accuracy of 96% and an AUC of 99.3%. In addition to biostatistical evaluation, in XAI-based SHAP results, increased expression levels of TSPYL5, ATP5E, CA9, NUP210, SLC37A1, ARIH1, PSMD7, UBQLN1, PRAME, and UBE2T (p <= 0.05) were found to be associated with an increased incidence of BC metastasis. Finally, decreased levels of expression of CACTIN, TGFB3, SCUBE2, ARL4D, OR1F1, ALDH4A1, PHF1, and CROCC (p <= 0.05) genes were also determined to increase the risk of metastasis in BC. Conclusion: The findings of this study may prevent disease progression and metastases and potentially improve clinical outcomes by recommending customized treatment approaches for BC patients.
Effect of high time under tension strength training on different muscular actions in the performance of runners: A randomized controlled trial
(Public Library Science, 2026) Martins, Julio Cesar de Carvalho; Fernandes, Matheus Santos de Sousa; Aidar, Felipe J.; de Carvalho e Silva, Gustavo Ivo; Alves Filho, Eder Magnus Almeida; Yagin, Fatma Hilal; de Souza, Raphael Fabricio
Background Strength training (ST) for runners is typically based on low volumes and high intensities. However, alternative approaches emphasizing higher volume and lower intensity, such as protocols with high time under tension (TUT), remain underexplored in this population. Notably, the effects of high-TUT may vary depending on the predominant type of muscle contraction.Objectives To analyze the impact of strength training with high time under tension on different muscle actions in performance-related variables in runners.Methods Thirty-four physically active young males were randomly divided into three groups: dynamic strength training, isometric strength training, and a control group. Over four weeks, the training groups performed two weekly sessions of bodyweight exercises with equalized time under tension per set (84 s), adjusted according to the type of contraction. The primary outcomes were performance in a 3000-meter time trial, peak torque, countermovement jump, neuromuscular fatigue, internal running load, ground contact time, and vertical oscillation. The data were analyzed using a two-way repeated measures analysis of variance.Results Peak torque increased by 13.3% in the dynamic group and 14.2% in the isometric group, compared with 2.5% in the control, with statistical significance only for the isometric group (p = 0.034, d = 1.12). Vertical jump height improved by 5.4% in the dynamic group and 4.1% in the isometric group compared with 0.7% in the control (p = 0.003, d = 1.54 and p = 0.030, d = 1.13, respectively).Conclusion In conclusion, high time under tension strength training, both dynamic and isometric, improved neuromuscular characteristics in runners. However, these adaptations did not translate into significant changes in running performance or running economy over the duration of the intervention.Trial registration This trial was registered in the Brazilian Clinical Trials Registry (ReBEC) with the identifier RBR-686kqdx.
Effect of high time under tension strength training on different muscular actions in the performance of runners: A randomized controlled trial [2]
(2026) Martins, Júlio César de Carvalho; Fernandes, Matheus Santos de Sousa; Aidar, Felipe J.; Silva, Gustavo Ivo de Carvalho E; Filho, Eder Magnus Almeida Alves; Yagin, Fatma Hilal; Souza, Raphael Fabricio de
BACKGROUND: Strength training (ST) for runners is typically based on low volumes and high intensities. However, alternative approaches emphasizing higher volume and lower intensity, such as protocols with high time under tension (TUT), remain underexplored in this population. Notably, the effects of high-TUT may vary depending on the predominant type of muscle contraction. OBJECTIVES: To analyze the impact of strength training with high time under tension on different muscle actions in performance-related variables in runners. METHODS: Thirty-four physically active young males were randomly divided into three groups: dynamic strength training, isometric strength training, and a control group. Over four weeks, the training groups performed two weekly sessions of bodyweight exercises with equalized time under tension per set (84 s), adjusted according to the type of contraction. The primary outcomes were performance in a 3000-meter time trial, peak torque, countermovement jump, neuromuscular fatigue, internal running load, ground contact time, and vertical oscillation. The data were analyzed using a two-way repeated measures analysis of variance. RESULTS: Peak torque increased by 13.3% in the dynamic group and 14.2% in the isometric group, compared with 2.5% in the control, with statistical significance only for the isometric group (p = 0.034, d = 1.12). Vertical jump height improved by 5.4% in the dynamic group and 4.1% in the isometric group compared with 0.7% in the control (p = 0.003, d = 1.54 and p = 0.030, d = 1.13, respectively). CONCLUSION: In conclusion, high time under tension strength training, both dynamic and isometric, improved neuromuscular characteristics in runners. However, these adaptations did not translate into significant changes in running performance or running economy over the duration of the intervention. TRIAL REGISTRATION: This trial was registered in the Brazilian Clinical Trials Registry (ReBEC) with the identifier RBR-686kqdx. Copyright: © 2026 Martins et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Effect of listening to preferred music at different frequencies during warmup on physical performance and psychophysiological responses in male athletes (vol 15, 36120 , 2025)
(Nature Portfolio, 2025) Jebabli, Nidhal; Boujabli, Manar; Ouergui, Ibrahim; Ouerghi, Nejmeddine; Yagin, Fatma Hilal; Yagin, Burak; Elkholi, Safaa M.
[Abstract Not Available]
Effects of a Mediterranean diet and structured exercise intervention on selected anthropometric, cardiovascular, and metabolic variables in physically inactive adults: a randomized controlled trial
(Frontiers Media Sa, 2025) Prieto-Gonzalez, Pablo; Yagin, Fatma Hilal; Alghannam, Abdullah F.; Canli, Umut
Objective: The study aims to determine whether the combined implementation of Mediterranean diet (MD) adherence and structured physical exercise contributes to improvements in body composition and cardiometabolic health indicators in a physically inactive but otherwise healthy adult population. Methods: A randomized controlled trial (RCTs) was conducted with 125 physically inactive adults (61 males, 64 females) aged 35-50 years, free from cardiovascular, metabolic, or musculoskeletal conditions. Participants were assigned to either an 8-week intervention group (n = 62: 30 males, 32 females) combining Mediterranean diet adherence and supervised combined training (three endurance and two resistance sessions per week) or a control group (n = 63: 31 males, 32 females) instructed to maintain habits. Anthropometric, cardiovascular, and metabolic variables were assessed pre- and post-intervention under standardized conditions. A 2 x 2 x 2 mixed-design ANOVA (group x sex x time) was conducted, with Tukey's post hoc tests applied when significant differences were found. Results: Significant differences over time, between sexes, and between groups were observed in anthropometric, cardiovascular, and metabolic variables. In the experimental group (EG), both men and women experienced significant reductions in body mass (BM), BMI, fat percentage, waist circumference, waist-to-hip ratio (WHR), systolic and diastolic blood pressure, heart rate, double product, glucose, and low-density lipoprotein (LDL) cholesterol from pre- to post-test (all p < 0.05, with effect sizes ranging from small to large). Lean mass increased significantly only in the EG, while high-density lipoprotein (HDL) levels improved predominantly in women. Men and women differed significantly in body mass, BMI, fat percentage, lean mass, waist circumference, waist-to-hip ratio, heart rate, double product, glucose, HDL, and triglycerides at both time points (all p < 0.001). The EG showed significantly greater improvements compared to the control group after the intervention (p < 0.001 for most variables), confirming the intervention's effectiveness. Conclusion: This study provides robust evidence that a lifestyle intervention combining Mediterranean diet adherence with structured physical exercise is an effective and feasible strategy to enhance cardiometabolic health in physically inactive adults. Its multicomponent nature and consistent benefits across sexes support its integration into preventive health programs. Further research is warranted to assess the sustainability of these outcomes and their generalizability to broader populations.
Effects of cooperative learning on students' learning outcomes in physical education: a meta-analysis
(Frontiers Media Sa, 2025) Boke, Hulusi; Aygun, Yalin; Tufekci, Sakir; Yagin, Fatma Hilal; Canpolat, Burak; Norman, Goktug; Ardigo, Luca Paolo
This meta-analysis examines the effect of Cooperative Learning (CL) interventions, compared to traditional instructional methods, on students' learning outcomes across affective, cognitive, physical, and social domains in physical education (PE). The review involved a comprehensive search of 12 databases in English, Spanish, and Turkish, with the last search conducted on June 2nd, 2024. Studies included were true experimental or quasi-experimental designs featuring direct CL interventions in PE, covering students of both genders from primary school to university levels. The standardized Cochrane methods were used to identify eligible records, collect and combine data, and assess the risk of bias. Comprehensive Meta-Analysis (CMA) v4 software package was used to yield a summary of quantitative results. Hedges's g was used as the effect size (ES) measure, calculated from pre- and post-tests in both experimental and control groups. Forty-three studies (comprising 60 reports) were initially included, but three studies were excluded as outliers, leaving 40 studies (56 reports) with a total of 3.985 participants for analysis. The random effects model revealed a moderate positive overall effect of CL interventions (ES = 0.459, 95% CI = [0.324, 0.592], p < 0.001), indicating that CL enhances PE students' learning across four domains. Subgroup analyses showed small to moderate ESs for affective (ES = 0.304), physical (ES = 0.471), cognitive (ES = 0.589), and social learning (ES = 0.612). Risk of bias was evaluated using Begg and Mazumdar's rank correlation, the classic fail-safe number, and a funnel plot, all indicating a low risk of bias. Methodological quality was assessed using the Medical Education Research Study Quality Instrument (MERSQI). The study was registered on PROSPERO (ID: CRD42024532607). This meta-analysis underscores the effectiveness of CL as a student-centered pedagogical model in PE, demonstrating its positive effect on various learning outcomes in the affective, cognitive, physical, and social domains. The findings provide instructive data and strategies for researchers, practitioners, and policymakers aiming to integrate, implement, or make context-specific adaptations of CL into educational processes, while ESs in the affective, physical, cognitive, and social learning domains provide domain-based implementation guidance for these stakeholders.
Effects of Swedish Massage at Different Times of the Day on Dynamic and Static Balance in Taekwondo Athletes
(Mdpi, 2024) Bayrakdaroglu, Serdar; Eken, Ozgur; Bayer, Ramazan; Yagin, Fatma Hilal; Kizilet, Tuba; Kayhan, Recep Fatih; Ardigo, Luca Paolo
The purpose of this study is to investigate the impact of different durations of Swedish massage on the static and dynamic balance at different times of the day in taekwondo athletes. Twelve taekwondo athletes who had been practicing on a regular basis for more than 5 years participated in this study. Taekwondo athletes completed static and dynamic balance tests either after a no-massage protocol (NMP), a five-minute massage protocol (5MMP), a ten-minute massage protocol (10MMP), or a fifteen-minute massage protocol (15MMP) two times a day in the morning (08:00-12:00) and in the evening (16:00-20:00), on non-consecutive days. The findings of this study suggest that the duration of the massage has a discernible impact on dynamic balance, particularly with regard to the right foot. Taekwondo athletes who received a 10MMP or 15MMP displayed significantly improved dynamic balance compared to those in the NMP. Importantly, these improvements were independent of the time of day when the massages were administered. It underscores the potential benefits of incorporating short-duration Swedish massages into taekwondo athletes' pre-competition routines to enhance dynamic balance. These findings highlight the potential benefits of incorporating short-duration Swedish massages into taekwondo athletes' pre-competition routines to enhance dynamic balance, a critical component of their performance, regardless of the time of day.
Enhancing Cognitive Health in Elderly Individuals: The Impact of Hatha Yoga on Attention, Memory, and Reasoning: A Randomized Controlled Trial
(Wiley, 2025) Oueslati, Rania; Souissi, Mohamed Abdelkader; Jarraya, Sana; Yagin, Fatma Hilal; Badicu, Georgian; Al-Hashem, Fahaid; Dahmen, Riadh
Background: Aging leads to physiological and psychological changes that compromise both mental and physical autonomy, as well as cognitive functions, thereby increasing the risk of anxiety and depression. The sedentary lifestyle typical of older individuals results in a deterioration of the overall quality of life and well-being. Objective: This study aims to evaluate the effectiveness of Hatha yoga in improving cognitive health among older adults. We will specifically examine the impact of this practice on attention, memory, and reasoning. Methods: The present study assesses the impact of Hatha yoga on attention, memorization, and reasoning in healthy older adults aged between 65 and 80 years. The study population comprises 45 healthy individuals (26 men and 19 women; 72.3 +/- 5.6 years) residing in a retirement home, divided into three groups: a yoga group (YOGA, n = 15) that participated in yoga sessions; a physical activity group (APS, n = 15) engaged in sports and physical activities sessions; and a control group (CONT, n = 15) that did not undertake any activities. The study spanned 24 sessions, with two sessions per week lasting 45 min each. Participants completed test sessions dedicated to evaluating attention, memory, and reasoning before (T0) and after (T1) 12 weeks. A two-way ANOVA was used to analyze the differences between groups and over time. Results: After the intervention sessions, the data showed that the YOGA group registered significantly greater improvements at T1 compared to that of T0 in all cognitive parameters (e.g., attention (p < 0.001, Hedges' g = 1.35), memory (p < 0.001, Hedges' g = 1.04), and reasoning (p < 0.001, Hedges' g = 1.82)). Furthermore, our results revealed a significant difference between the YOGA group and both the APS (p < 0.001) and CONT (p < 0.01) groups for the attention and reasoning parameters at T1. Conclusions: This study underscores the potential of Hatha yoga to enhance the mental well-being of the elderly, suggesting significant benefits for cognitive well-being in this population. Trial Registration: Pan African Clinical Trials Registry: PACTR202405804830163.
Explainable Boosting Machines Identify Key Metabolomic Biomarkers in Rheumatoid Arthritis
(Mdpi, 2025) Yagin, Fatma Hilal; Colak, Cemil; Algarni, Abdulmohsen; Algarni, Ali; Al-Hashem, Fahaid; Ardigo, Luca Paolo
Background and Objectives: Rheumatoid arthritis (RA) is a chronic autoimmune disease characterised by joint inflammation and pain. Metabolomics approaches, which are high-throughput profiling of small molecule metabolites in plasma or serum in RA patients, have so far provided biomarker discovery in the literature for clinical subgroups, risk factors, and predictors of treatment response using classical statistical approaches or machine learning models. Despite these recent developments, an explainable artificial intelligence (XAI)-based methodology has not been used to identify RA metabolomic biomarkers and distinguish patients with RA. This study constructed a XAI-based EBM model using global plasma metabolomics profiling to identify metabolites predictive of RA patients and to develop a classification model that can distinguish RA patients from healthy controls. Materials and Methods: Global plasma metabolomics data were analysed from RA patients (49 samples) and healthy individuals (10 samples). SMOTE technique was used for class imbalance in data preprocessing. EBM, LightGBM, and AdaBoost algorithms were applied to generate a discriminatory model between RA and controls. Comprehensive performance metrics were calculated, and the interpretability of the optimal model was assessed using global and local feature descriptions. Results: A total of 59 samples were analysed, 49 from RA patients, and 10 from healthy subjects. The EBM generated better results than LightGBM and AdaBoost by attaining an AUC of 0.901 (95% CI: 0.847-0.955) with 87.8% sensitivity which helps prevent false negative early RA diagnosis. The primary biomarkers EBM-based XAI identified were N-acetyleucine, pyruvic acid, and glycerol-3-phosphate. EBM global explanation analysis indicated that elevated pyruvic acid levels were significantly correlated with RA, whereas N-acetyleucine exhibited a nonlinear relationship, implying possible protective effects at specific concentrations. Conclusions: This study underscores the promise of XAI and evidence-based medicine methodology in developing biomarkers for RA through metabolomics. The discovered metabolites offer significant insights into RA pathophysiology and may function as diagnostic biomarkers or therapeutic targets. Incorporating EBM methodologies integrated with XAI improves model transparency and increases the therapeutic applicability of predictive models for RA diagnosis/management. Furthermore, the transparent structure of the EBM model empowers clinicians to understand and verify the reasoning behind each prediction, thereby fostering trust in AI-assisted decision-making and facilitating the integration of metabolomic insights into routine clinical practice.
HearteXplain: explainable prediction of acute heart failure and identification of hematologic biomarkers using EBMs and Morris sensitivity analysis
(Nature Portfolio, 2025) Yagin, Fatma Hilal; Gormez, Yasin; Algarni, Abdulmohsen; Al-Hashem, Fahaid; Dutta, Ashit Kumar; Aghaei, Mohammadreza
Hematological biomarkers have emerged as powerful tools in diagnosing Acute Heart Failure (AHF). This study introduces a novel diagnostic framework that integrates Explainable Artificial Intelligence (XAI) with Morris Sensitivity Analysis (MSA) to enhance both the interpretability and performance of machine learning models in AHF detection. A dataset consisting of 425 AHF patients and 430 controls was analyzed using eight machine learning models, including XGBoost, Histogram-based Gradient Boosting (histGB), Explainable Boosting Machine (EBM), and Random Forest. Model performance was evaluated through metrics such as AUC, accuracy, precision, recall, and Brier score. Hyperparameters were optimized via Bayesian optimization. Feature importance was assessed using MSA to identify variables with the highest predictive influence. The histGB model achieved the highest performance with an AUC of 87.93%. Both MSA and EBM consistently identified PDW, RDW-CV, NEU, NEU/LY ratio, age, and WBC as top predictive features across multiple models. These hematological markers demonstrated strong potential for early diagnosis and risk stratification in AHF patients. This study presents a clinically relevant, interpretable, and cost-effective diagnostic strategy that combines XAI with MSA for AHF prediction. The framework enhances clinical trust and provides a pathway toward personalized treatment by identifying accessible hematological biomarkers. The integration of explainability into AI models improves their transparency and applicability in real-world clinical settings.
Identification of a Novel Lipidomic Biomarker for Hepatocyte Carcinoma Diagnosis: Advanced Boosting Machine Learning Techniques Integrated with Explainable Artificial Intelligence
(Mdpi, 2025) Yagin, Fatma Hilal; Colak, Cemil; Al-Hashem, Fahaid; Alzakari, Sarah A.; Alhussan, Amel Ali; Aghaei, Mohammadreza
Background: Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality worldwide, often diagnosed at late stages due to the limited sensitivity of current screening tools. This study explores whether blood-based lipidomic profiling, combined with explainable artificial intelligence (XAI), can improve early and interpretable detection of HCC. Methods: We analyzed lipidomic data from 219 HCC patients and 219 matched healthy controls using liquid chromatography-mass spectrometry. An Explainable Boosting Machine (EBM) was employed to identify discriminatory lipid biomarkers and was compared against several standard machine learning algorithms. Results: The EBM model achieved superior performance with 87.0% accuracy, 87.7% sensitivity, 86.3% specificity, and an AUC of 91.8%, outperforming other models. Key lipid biomarkers identified included specific phosphatidylcholines (PC 38:2, PC 40:4), sphingomyelins (SM d40:2 B), and lysophosphatidylcholines (LPC 18:2), which exhibited significant alterations in HCC patients and highlighted disruptions in sphingolipid metabolism. Conclusions: Integration of lipidomics with explainable machine learning offers a powerful, transparent approach for HCC biomarker discovery, achieving high diagnostic accuracy while providing biological insights. This strategy holds promise for developing non-invasive, clinically interpretable screening tools to improve early detection of liver cancer.
Interpretable Machine Learning for Serum-Based Metabolomics in Breast Cancer Diagnostics: Insights from Multi-Objective Feature Selection-Driven LightGBM-SHAP Models
(Mdpi, 2025) Guldogan, Emek; Yagin, Fatma Hilal; Ucuzal, Hasan; Alzakari, Sarah A.; Alhussan, Amel Ali; Ardigo, Luca Paolo
Background and Objectives: Breast cancer accounts for 12.5% of all new cancer cases in women worldwide. Early detection significantly improves survival rates, but traditional biomarkers like CA 15-3 and HER2 lack sensitivity and specificity, particularly for early-stage disease. Advances in metabolomics and machine learning, particularly explainable artificial intelligence (XAI), offer new opportunities for identifying robust biomarkers and improving diagnostic accuracy. This study aimed to identify and validate serum-based metabolic biomarkers for breast cancer using advanced metabolomic profiling techniques and a Light Gradient Boosting Machine (LightGBM) model. Additionally, SHapley Additive exPlanations (SHAP) were applied to enhance model interpretability and biological insight. Materials and Methods: The study included 103 breast cancer patients and 31 healthy controls. Serum samples underwent liquid and gas chromatography-time-of-flight mass spectrometry (LC-TOFMS and GC-TOFMS). Mutual Information (MI), Sparse Partial Least Squares (sPLS), Boruta, and Multi-Objective Feature Selection (MOFS) approaches were applied to the data for biomarker discovery. LightGBM, AdaBoost, and Random Forest were employed for classification and to identify class imbalance with the Synthetic Minority Oversampling Technique (SMOTE). SHAP analysis ranked metabolites based on their contribution to model predictions. Results: Compared to other feature selection approaches, the MOFS approach was more robust in terms of predictive performance, and metabolites identified by this method were used in subsequent analyses for biomarker discovery. LightGBM outperformed the AdaBoost and Random Forest models, achieving 86.6% accuracy, 89.1% sensitivity, 84.2% specificity, and an F1-score of 87.0%. SHAP analysis identified 2-Aminobutyric acid, choline, and coproporphyrin as the most influential metabolites, with dysregulation of these markers associated with breast cancer risk. Conclusions: This study is among the first to integrate SHAP explainability with metabolomic profiling, bridging computational predictions and biological insights for improved clinical adoption. This study demonstrates the effectiveness of combining metabolomics with XAI-driven machine learning for breast cancer diagnostics. The identified biomarkers not only improve diagnostic accuracy but also reveal critical metabolic dysregulations associated with disease progression.
Is environmental enrichment effective in modulating autophagy markers in the brain exposed to adverse conditions? A systematic review
(Frontiers Media Sa, 2025) Silva, Clarice Beatriz Goncalves; de Sousa Fernandes, Matheus Santos; Cerqueira, Debora Dantas Nucci; Santos, Gabriela Carvalho Jurema; Yagin, Fatma Hilal; Aygun, Yalin; Tabnjh, Abedelmalek Kalefh
Autophagy is a key regulator of cellular homeostasis and neuronal survival, particularly under adverse physiological conditions. Environmental enrichment (EE), a non-pharmacological intervention providing enhanced sensory, cognitive, and motor stimulation, may modulate autophagic processes in the brain. This systematic review aimed to synthesize preclinical findings on the effects of EE on autophagy markers in rodent models subjected to diverse adverse conditions. A literature search across PubMed, Scopus, ScienceDirect, and embase yielded eight eligible studies meeting inclusion criteria. EE was found to be generally associated with upregulation of key autophagic markers such as Beclin-1, LC3-II/LC3-I ratio, cathepsins, p62, p-TFEB, and LAMP-1 across brain regions including the cortex, hippocampus, and penumbral area. However, reductions in some markers were also observed, indicating that the modulatory effects of EE are context-dependent and may vary with disease model, brain region, or EE protocol duration. These findings suggest that EE holds promise as an adjunctive strategy to modulate autophagy and mitigate neurodegeneration, though heterogeneity in study design and outcomes warrants caution during interpretation. Further mechanistic and sex-specific studies are needed to clarify the therapeutic relevance of EE-induced autophagic modulation.
Leveraging Explainable Automated Machine Learning (AutoML) and Metabolomics for Robust Diagnosis and Pathophysiological Insights in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS)
(Mdpi, 2025) Yagin, Fatma Hilal; Colak, Cemil; Al-Hashem, Fahaid; Alzakari, Sarah A.; Alhussan, Amel Ali; Aghaei, Mohammadreza
Background/Objectives: Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is a debilitating complex disease with an elusive etiology, lacking objective diagnostic biomarkers. This study leverages advanced Automated Machine Learning (AutoML) to analyze plasma metabolomic and lipidomic profiles for the purpose of ME/CFS detection. Methods: We utilized a publicly available dataset comprising 888 metabolic features from 106 ME/CFS patients and 91 matched controls. Three AutoML frameworks-TPOT, Auto-Sklearn, and H2O AutoML-were benchmarked under identical time constraints. Univariate ROC and PLS-DA analyses with cross-validation, permutation testing, and VIP-based feature selection were applied to standardized, log-transformed omics data to identify significant discriminatory metabolites/lipids and assess their intercorrelations. Results: TPOT significantly outperformed its counterparts, achieving an area under the curve (AUC) of 92.1%, accuracy of 87.3%, sensitivity of 85.8%, and specificity of 89.0%. The PLS-DA model revealed a moderate but statistically significant discrimination between ME/CFS and controls. Explainable artificial intelligence (XAI) via SHAP analysis of the optimal TPOT model identified key metabolites implicating dysregulated pathways in mitochondrial energy metabolism (succinic acid, pyruvic acid, leucine), chronic inflammation (prostaglandin D2, 11,12-EET), gut-brain axis communication (glycocholic acid), and cell membrane integrity (pc(35:2)a). Conclusions: Our results demonstrate that TPOT-derived models not only provide a highly accurate and robust diagnostic tool but also yield biologically interpretable insights into the pathophysiology of ME/CFS, highlighting its potential for clinical decision support and elucidating novel therapeutic targets.
Machine Learning Classification of Cognitive Status in Community-Dwelling Sarcopenic Women: A SHAP-Based Analysis of Physical Activity and Anthropometric Factors
(Mdpi, 2025) Gormez, Yasin; Yagin, Fatma Hilal; Aygun, Yalin; Alzakari, Sarah A.; Alhussan, Amel Ali; Aghaei, Mohammadreza
Background and Objectives: Sarcopenia, characterized by progressive loss of skeletal muscle mass and function, has increasingly been recognized not only as a physical health concern but also as a potential risk factor for cognitive decline. This study investigates the application of machine learning algorithms to classify cognitive status based on Mini-Mental State Examination (MMSE) scores in community-dwelling sarcopenic women. Materials and Methods: A dataset of 67 participants was analyzed, with MMSE scores categorized into severe (<= 17) and mild (>17) cognitive impairment. Eight classification models-MLP, CatBoost, LightGBM, XGBoost, Random Forest (RF), Gradient Boosting (GB), Logistic Regression (LR), and AdaBoost-were evaluated using a repeated holdout strategy over 100 iterations. Hyperparameter optimization was performed via Bayesian optimization, and model performance was assessed using metrics including weighted F1-score (w_f1), accuracy, precision, recall, PR-AUC, and ROC-AUC. Results: Among the models, CatBoost achieved the highest w_f1 (87.05 +/- 2.85%) and ROC-AUC (90 +/- 5.65%), while AdaBoost and GB showed superior PR-AUC scores (92.49% and 91.88%, respectively), indicating strong performance in handling class imbalance and threshold sensitivity. SHAP (SHapley Additive exPlanations) analysis revealed that moderate physical activity (moderatePA minutes), walking days, and sitting time were among the most influential features, with higher physical activity associated with reduced risk of cognitive impairment. Anthropometric factors such as age, BMI, and weight also contributed significantly. Conclusions: The results highlight the effectiveness of boosting-based models in capturing complex patterns in clinical data and provide interpretable evidence supporting the role of modifiable lifestyle factors in cognitive health. These findings suggest that machine learning, combined with explainable AI, can enhance risk assessment and inform targeted interventions for cognitive decline in older women.

Yazar "Yagin, Fatma Hilal" seçeneğine göre listele

Sayfa Başına Sonuç

Sıralama seçenekleri