Khadijeh Moulaei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Khadijeh Moulaei has a total of 84 co-authors. Here are the top co-authors by number of shared publications:
- Kambiz Bahaadinbeigy
Digital Health Team, The Australian College of Rural and Remote Medicine Brisbane Queensland Australia
12 shared publications Fatemeh Dinari
Institute for Futures Studies in Health, Kerman University of Medical Sciences
5 shared publications- Fatemeh Dinari
Medical Informatics Research Center, Institute for Futures Studies in Health Kerman University of Medical Sciences Kerman Iran
4 shared publications - Hadi Kazemi-Arpanahi
Department of Health Information Technology, Abadan University of Medical Sciences, Abadan, Iran
4 shared publications Kambiz Bahaadinbeigy
Medical Informatics Research Center Institute for Futures Studies in Health Kerman University of Medical Sciences Kerman Iran
4 shared publicationsReza Moulaei
Department of Orthopedics, School of Medicine Valiasr Hospital, Arak University of Medical Sciences, Arak, Iran
3 shared publications
Latest Publications
- Open Access
- 9 October 2026
- Published by SAGE Publications in Journal of International Medical Research
- Vol. 54 (10)
Navigating lupus with ease: A patient-driven mobile application for improved self-management
Background: Lupus can affect various aspects of patients’ lives, including family relationships, daily activities, work, and independence, because of inflammation and disease-related complications. Self-management applications are among the potential solutions for controlling and managing these complications. Therefore, this study aimed to design and evaluate a mobile-based application for self-management among patients with lupus. Methods: This study was conducted in two phases. In the first phase, a systematic review was conducted to identify the information needs of patients and the required capabilities of the application. The identified information needs and capabilities were then confirmed by 12 patients with lupus and six rheumatologists. In the second phase, based on the information and capabilities confirmed in the previous phase, an application prototype was developed and evaluated. To evaluate the application, eight patients with lupus used the application for 1 week. SPSS version 23 was used to analyze the data obtained from both phases. Results: According to the opinions of patients and physicians, 61 of 70 information items and capabilities met the predefined 75% inclusion threshold and were considered for application design, whereas nine items that scored below this threshold were excluded. The application included four domains: user profile, lifestyle, disease management and control, and application capabilities. Patients rated the usability of the application as good, with a mean score of 7.5 out of 9. Conclusion: The application developed in this study demonstrated good usability and has the potential to support self-management and self-care among patients with lupus. However, longer-term studies involving larger and more diverse samples are needed to assess sustained use and potential clinical impact.
- 9 October 2026
- Published by SAGE Publications in Journal of International Medical Research
- Vol. 54 (10)
Open AccessNavigating lupus with ease: A patient-driven mobile application for improved self-management
Background: Lupus can affect various aspects of patients’ lives, including family relationships, daily activities, work, and independence, because of inflammation and disease-related complications. Self-management applications are among the potential solutions for controlling and managing these complications. Therefore, this study aimed to design and evaluate a mobile-based application for self-management among patients with lupus. Methods: This study was conducted in two phases. In the first phase, a systematic review was conducted to identify the information needs of patients and the required capabilities of the application. The identified information needs and capabilities were then confirmed by 12 patients with lupus and six rheumatologists. In the second phase, based on the information and capabilities confirmed in the previous phase, an application prototype was developed and evaluated. To evaluate the application, eight patients with lupus used the application for 1 week. SPSS version 23 was used to analyze the data obtained from both phases. Results: According to the opinions of patients and physicians, 61 of 70 information items and capabilities met the predefined 75% inclusion threshold and were considered for application design, whereas nine items that scored below this threshold were excluded. The application included four domains: user profile, lifestyle, disease management and control, and application capabilities. Patients rated the usability of the application as good, with a mean score of 7.5 out of 9. Conclusion: The application developed in this study demonstrated good usability and has the potential to support self-management and self-care among patients with lupus. However, longer-term studies involving larger and more diverse samples are needed to assess sustained use and potential clinical impact.
- Open Access
- 28 May 2026
- Published by Springer Nature in BMC Health Services Research
- Vol. 26 (1)
In-person or tele-orthopaedic care? A mixed-methods study exploring parents’ perspectives on the advantages and challenges of both approaches care for children with cerebral palsy
Children with cerebral palsy (CP) often require ongoing orthopaedic care to address musculoskeletal issues and improve mobility. However, access to special
- 28 May 2026
- Published by Springer Nature in BMC Health Services Research
- Vol. 26 (1)
Open AccessIn-person or tele-orthopaedic care? A mixed-methods study exploring parents’ perspectives on the advantages and challenges of both approaches care for children with cerebral palsy
Children with cerebral palsy (CP) often require ongoing orthopaedic care to address musculoskeletal issues and improve mobility. However, access to special
- Open Access
- 20 May 2026
- Published by Wiley in Health Science Reports
- Vol. 9 (5) , e72545
Geographical Distribution of Patients With Chronic Kidney Disease Receiving Hemodialysis Care in Northwest Iran: A Retrospective Cross‐Sectional Study
Background and Aims: Managing chronic kidney disease (CKD) care, particularly for patients requiring hemodialysis, presents significant challenges due to the uneven geographical distribution of cases, which complicates equitable resource allocation. Geographic Information Systems (GIS) can help identify spatial patterns and clusters, thereby supporting more efficient healthcare planning. This study aimed to investigate the spatial distribution of CKD cases requiring hemodialysis in West Azerbaijan Province, Iran. Methods: This retrospective cross‐sectional descriptive study analyzed hemodialysis patient data collected during January–December 2021. Data from 1709 patients undergoing maintenance hemodialysis were obtained from all 19 governmental and 2 private hemodialysis centers operating at the provincial level. As all registered dialysis centers were included, the data set was considered representative of the provincial hemodialysis population. Records with incomplete key variables were excluded from analysis. Spatial prevalence patterns were assessed using spatial autocorrelation techniques, including the global Moran's I and local Moran's I indices, to identify nonrandom spatial distributions. Results: The prevalence of hemodialysis in West Azerbaijan Province was 52.33 per 100,000 population. The global Moran's I index was 0.2706 ( Z = 1.82, p < 0.10), indicating a clustered spatial pattern with positive spatial autocorrelation. Local Moran's I analysis revealed no high–high clusters of hemodialysis prevalence. However, two significant low–low clusters (cold spots) were identified in the cities of Showt and Mako, while a high–low cluster was observed in Qarah Zia ol Din. Conclusion: The findings highlight spatial inequalities in hemodialysis distribution and underscore the importance of geographically targeted resource allocation and public health strategies to improve CKD care delivery in the region.
- 20 May 2026
- Published by Wiley in Health Science Reports
- Vol. 9 (5) , e72545
Open AccessGeographical Distribution of Patients With Chronic Kidney Disease Receiving Hemodialysis Care in Northwest Iran: A Retrospective Cross‐Sectional Study
Background and Aims: Managing chronic kidney disease (CKD) care, particularly for patients requiring hemodialysis, presents significant challenges due to the uneven geographical distribution of cases, which complicates equitable resource allocation. Geographic Information Systems (GIS) can help identify spatial patterns and clusters, thereby supporting more efficient healthcare planning. This study aimed to investigate the spatial distribution of CKD cases requiring hemodialysis in West Azerbaijan Province, Iran. Methods: This retrospective cross‐sectional descriptive study analyzed hemodialysis patient data collected during January–December 2021. Data from 1709 patients undergoing maintenance hemodialysis were obtained from all 19 governmental and 2 private hemodialysis centers operating at the provincial level. As all registered dialysis centers were included, the data set was considered representative of the provincial hemodialysis population. Records with incomplete key variables were excluded from analysis. Spatial prevalence patterns were assessed using spatial autocorrelation techniques, including the global Moran's I and local Moran's I indices, to identify nonrandom spatial distributions. Results: The prevalence of hemodialysis in West Azerbaijan Province was 52.33 per 100,000 population. The global Moran's I index was 0.2706 ( Z = 1.82, p < 0.10), indicating a clustered spatial pattern with positive spatial autocorrelation. Local Moran's I analysis revealed no high–high clusters of hemodialysis prevalence. However, two significant low–low clusters (cold spots) were identified in the cities of Showt and Mako, while a high–low cluster was observed in Qarah Zia ol Din. Conclusion: The findings highlight spatial inequalities in hemodialysis distribution and underscore the importance of geographically targeted resource allocation and public health strategies to improve CKD care delivery in the region.
- Open Access
- 28 April 2026
- Published by Wiley in Health Science Reports
- Vol. 9 (5) , e72261
How Can Large Language Models Improve Health Research Skills: A Perspective Study
Background and Aims: Health researchers often struggle with challenges such as translating complex texts, interpreting nuanced information, conducting detailed statistical analyses, and more. Artificial Intelligence (AI) models, by providing easy access to information, offer a promising solution to these challenges. This study aims to examine university students' perspectives on the impact of AI models in overcoming these challenges. Methods: This survey study was conducted in 2023 by using a researcher‐developed questionnaire. A total of 445 students from the Kerman University of Medical Sciences participated and completed the questionnaire. Descriptive statistics and linear regression were used to analyze the relationship between independent variables and the impact of AI models on health research. Open‐ended responses were coded and stored in MAXQDA. Results: Sixty percent of students reported moderate or greater familiarity with AI models, while only 52% had moderate or greater usage of these tools. Over 70% considered the associated risks high. Participants rated AI models most favorably for “Translation of an article” (Mean = 3.33, SD = 1.07), “Article rephrasing” (Mean = 3.33, SD = 1.07), and “Reference search” (Mean = 3.27, SD = 1.14), with the lowest ratings for “Article submissions” (Mean = 3.10, SD = 1.01), “Data analysis” (Mean = 3.23, SD = 1.01), and “Proposal submissions” (Mean = 3.23, SD = 0.91), respectively. The most important applications of AI models were noted in “Translation of research” (Mean = 3.44), “Grammer checking” (Mean = 3.46, SD = 1.12), and “Text editing” (Mean = 3.35, SD = 1.14), while the least were in “Correspondence with journal editors” (Mean = 2.99, SD = 1.20), “Adherence to journal guidelines” (Mean = 3.07, SD = 1.17), and “Response to the reviewers' comments” (Mean = 3.12, SD = 1.13). “Providing false information” ( n = 34), “Failure to apply legal and ethical standards in research” ( n = 10), and a “Loss of ability and creativity among researchers” ( n = 9) were the most significant risks of using AI models in health research. Conclusions: AI models show promise for improving research efficiency and quality, but further efforts are necessary to mitigate perceived risks and encourage broader adoption.
- 28 April 2026
- Published by Wiley in Health Science Reports
- Vol. 9 (5) , e72261
Open AccessHow Can Large Language Models Improve Health Research Skills: A Perspective Study
Background and Aims: Health researchers often struggle with challenges such as translating complex texts, interpreting nuanced information, conducting detailed statistical analyses, and more. Artificial Intelligence (AI) models, by providing easy access to information, offer a promising solution to these challenges. This study aims to examine university students' perspectives on the impact of AI models in overcoming these challenges. Methods: This survey study was conducted in 2023 by using a researcher‐developed questionnaire. A total of 445 students from the Kerman University of Medical Sciences participated and completed the questionnaire. Descriptive statistics and linear regression were used to analyze the relationship between independent variables and the impact of AI models on health research. Open‐ended responses were coded and stored in MAXQDA. Results: Sixty percent of students reported moderate or greater familiarity with AI models, while only 52% had moderate or greater usage of these tools. Over 70% considered the associated risks high. Participants rated AI models most favorably for “Translation of an article” (Mean = 3.33, SD = 1.07), “Article rephrasing” (Mean = 3.33, SD = 1.07), and “Reference search” (Mean = 3.27, SD = 1.14), with the lowest ratings for “Article submissions” (Mean = 3.10, SD = 1.01), “Data analysis” (Mean = 3.23, SD = 1.01), and “Proposal submissions” (Mean = 3.23, SD = 0.91), respectively. The most important applications of AI models were noted in “Translation of research” (Mean = 3.44), “Grammer checking” (Mean = 3.46, SD = 1.12), and “Text editing” (Mean = 3.35, SD = 1.14), while the least were in “Correspondence with journal editors” (Mean = 2.99, SD = 1.20), “Adherence to journal guidelines” (Mean = 3.07, SD = 1.17), and “Response to the reviewers' comments” (Mean = 3.12, SD = 1.13). “Providing false information” ( n = 34), “Failure to apply legal and ethical standards in research” ( n = 10), and a “Loss of ability and creativity among researchers” ( n = 9) were the most significant risks of using AI models in health research. Conclusions: AI models show promise for improving research efficiency and quality, but further efforts are necessary to mitigate perceived risks and encourage broader adoption.
- Open Access
- 2 March 2026
- Published by Wiley in Health Science Reports
- Vol. 9 (3)
The Impact of Arbaeen Long‐Walk on the Health of Pilgrims: A Qualitative Study
Background and Aims: The Arbaeen long walk (ALW) is an annual religious pilgrimage in which over 20 million Muslims walk more than 80 km between Najaf and Karbala in Iraq. While this event may affect participants' health, its impact has not been comprehensively studied. This study aimed to explore the perceived effects of the Arbaeen walk on individuals' health across multiple dimensions. Methods: In this qualitative study, 19 university students who participated in ALW were interviewed using a semi‐structured guide. Content analysis was applied to the interview data using MAXQDA 2018. The study followed the Consolidated Criteria for Reporting Qualitative Research (COREQ) to ensure methodological rigor. Results: A total of 264 initial codes were identified, which were consolidated into 78 unique codes, including 52 positive and 24 negative outcomes. These groups comprised 78 unique codes, with 52 codes representing positive outcomes of the Arbaeen walk and 24 codes representing negative outcomes. Based on frequency, the most significant positive outcomes of the Arbaeen walk were “Peace of mind and heart,” “Cultural exchange with other countries,” and “Strengthening the relationship between God and individuals.” In terms of negative outcomes, the code frequency highlighted the prevalence of “Outbreak of infectious diseases,” “Physical fatigue,” and “Excessive waste generation.” Conclusion: Participation in the Arbaeen walk is associated with diverse positive effects on physical, mental, and social well‐being, although some negative health outcomes were also reported. These findings highlight the broad scope of health impacts related to this pilgrimage and underscore the need for further investigation into its effects.
- 2 March 2026
- Published by Wiley in Health Science Reports
- Vol. 9 (3)
Open AccessThe Impact of Arbaeen Long‐Walk on the Health of Pilgrims: A Qualitative Study
Background and Aims: The Arbaeen long walk (ALW) is an annual religious pilgrimage in which over 20 million Muslims walk more than 80 km between Najaf and Karbala in Iraq. While this event may affect participants' health, its impact has not been comprehensively studied. This study aimed to explore the perceived effects of the Arbaeen walk on individuals' health across multiple dimensions. Methods: In this qualitative study, 19 university students who participated in ALW were interviewed using a semi‐structured guide. Content analysis was applied to the interview data using MAXQDA 2018. The study followed the Consolidated Criteria for Reporting Qualitative Research (COREQ) to ensure methodological rigor. Results: A total of 264 initial codes were identified, which were consolidated into 78 unique codes, including 52 positive and 24 negative outcomes. These groups comprised 78 unique codes, with 52 codes representing positive outcomes of the Arbaeen walk and 24 codes representing negative outcomes. Based on frequency, the most significant positive outcomes of the Arbaeen walk were “Peace of mind and heart,” “Cultural exchange with other countries,” and “Strengthening the relationship between God and individuals.” In terms of negative outcomes, the code frequency highlighted the prevalence of “Outbreak of infectious diseases,” “Physical fatigue,” and “Excessive waste generation.” Conclusion: Participation in the Arbaeen walk is associated with diverse positive effects on physical, mental, and social well‐being, although some negative health outcomes were also reported. These findings highlight the broad scope of health impacts related to this pilgrimage and underscore the need for further investigation into its effects.
- Open Access
- 28 January 2026
- Published by SAGE Publications in INQUIRY: The Journal of Health Care Organization, Provision, and Financing
- Vol. 63
Assessing the Impact of Virtual Reality, Augmented Reality, and Video Games on Improving Post-Traumatic Stress Disorder Symptoms: A Systematic Review and Meta-Analysis
Post-traumatic stress disorder (PTSD) is often debilitating, with current treatments limited by low adherence, high costs, and accessibility issues. Innovative technologies such as virtual reality (VR), augmented reality (AR), and therapeutic video games provide immersive environments that may improve treatment outcomes. This systematic review and meta-analysis evaluated the efficacy of these approaches and explored their potential advantages over traditional methods. A comprehensive search of PubMed, PsycINFO, CINAHL, Web of Science, and Cochrane identified relevant studies. Two reviewers independently screened articles, extracted data, and assessed quality using the Mixed Methods Appraisal Tool (MMAT). A random-effects model was used to calculate pooled effect sizes (Hedges’ g), and heterogeneity was evaluated with the Q test and I 2 statistic. Publication bias was examined with funnel plots, Egger’s, and Begg’s tests. Analyses were performed in Stata version 17.0. From 480 records, 21 studies were included in the review and 12 in the meta-analysis. VR-based treatments yielded a pooled effect size of –0.35 (95% CI [–0.57, –0.13]), indicating a small-to-moderate reduction in PTSD symptoms. The effect was statistically significant ( z = –3.13, P < .01), with moderate heterogeneity ( I 2 = 46.28%, P = .03). Funnel plots and statistical tests suggested minimal publication bias. Meta-regression showed no moderating effect of gender. Subgroup analyses indicated significant benefits in male-only samples, participants aged 20 to 30 and over 40, and studies with follow-up periods ≤7 months. Larger effects were observed in studies with 15 to 30 participants. VR, AR, and video game interventions significantly reduce PTSD symptoms and may enhance accessibility and engagement compared to traditional treatments. These findings support the integration of immersive technologies into therapeutic practice to improve outcomes for individuals with PTSD.
- 28 January 2026
- Published by SAGE Publications in INQUIRY: The Journal of Health Care Organization, Provision, and Financing
- Vol. 63
Open AccessAssessing the Impact of Virtual Reality, Augmented Reality, and Video Games on Improving Post-Traumatic Stress Disorder Symptoms: A Systematic Review and Meta-Analysis
Post-traumatic stress disorder (PTSD) is often debilitating, with current treatments limited by low adherence, high costs, and accessibility issues. Innovative technologies such as virtual reality (VR), augmented reality (AR), and therapeutic video games provide immersive environments that may improve treatment outcomes. This systematic review and meta-analysis evaluated the efficacy of these approaches and explored their potential advantages over traditional methods. A comprehensive search of PubMed, PsycINFO, CINAHL, Web of Science, and Cochrane identified relevant studies. Two reviewers independently screened articles, extracted data, and assessed quality using the Mixed Methods Appraisal Tool (MMAT). A random-effects model was used to calculate pooled effect sizes (Hedges’ g), and heterogeneity was evaluated with the Q test and I 2 statistic. Publication bias was examined with funnel plots, Egger’s, and Begg’s tests. Analyses were performed in Stata version 17.0. From 480 records, 21 studies were included in the review and 12 in the meta-analysis. VR-based treatments yielded a pooled effect size of –0.35 (95% CI [–0.57, –0.13]), indicating a small-to-moderate reduction in PTSD symptoms. The effect was statistically significant ( z = –3.13, P < .01), with moderate heterogeneity ( I 2 = 46.28%, P = .03). Funnel plots and statistical tests suggested minimal publication bias. Meta-regression showed no moderating effect of gender. Subgroup analyses indicated significant benefits in male-only samples, participants aged 20 to 30 and over 40, and studies with follow-up periods ≤7 months. Larger effects were observed in studies with 15 to 30 participants. VR, AR, and video game interventions significantly reduce PTSD symptoms and may enhance accessibility and engagement compared to traditional treatments. These findings support the integration of immersive technologies into therapeutic practice to improve outcomes for individuals with PTSD.
- 21 January 2026
- Published by Elsevier in Health Policy and Technology
- Vol. 15 (3)
Telemedicine-based geriatric assessment (tele-GA): A scoping review on applications, benefits and challenges
- 21 January 2026
- Published by Elsevier in Health Policy and Technology
- Vol. 15 (3)
Telemedicine-based geriatric assessment (tele-GA): A scoping review on applications, benefits and challenges
- Open Access
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in Healthcare Technology Letters
- Vol. 13 (1)
Empowering People With Haemophilia: A Mobile Solution for Self‐Management
Background and Aims: Haemophilia is a congenital bleeding disorder requiring extensive self‐management. Mobile applications can support these processes. This study aimed to design and evaluate a mobile application to facilitate self‐management in persons with haemophilia. Methods: The study was conducted in two phases. First, a literature review was performed in PubMed, Web of Science, and Scopus, and searches were made in Google Play, Apple Store, and the websites of the World Federation of Hemophilia and Hemophilia Federation of America to identify information‐educational needs and necessary capabilities for app design. A questionnaire based on these needs was then provided to physicians and patients for approval. In the second phase, the application prototype was designed according to the approved information‐educational needs and necessary capabilities and evaluated for usability using the standard Questionnaire for User Interface Satisfaction with 17 patients. Data were analysed using SPSS 20. Results: Sixty‐two information‐educational needs and capabilities with scores above 50% were included in the design. The designed application can provide the conditions for recording demographic information, medical history, treatment information, bleeding information, and injuries and accidents and provide the education on self‐management for patients. Moreover, it has the capabilities to set various reminders (taking medication, rehabilitation‐physical activities, and appointments with the therapist), set the treatment plan, send reports to the therapist, communicate with other persons with haemophilia and therapists, and record notes. Patients evaluated the application's usability at a “good” level. Conclusion: The haemophilia self‐management app can enhance patients’ quality of life by addressing their educational needs and supporting self‐management.
- 1 January 2026
- Published by Institution of Engineering and Technology (IET) in Healthcare Technology Letters
- Vol. 13 (1)
Open AccessEmpowering People With Haemophilia: A Mobile Solution for Self‐Management
Background and Aims: Haemophilia is a congenital bleeding disorder requiring extensive self‐management. Mobile applications can support these processes. This study aimed to design and evaluate a mobile application to facilitate self‐management in persons with haemophilia. Methods: The study was conducted in two phases. First, a literature review was performed in PubMed, Web of Science, and Scopus, and searches were made in Google Play, Apple Store, and the websites of the World Federation of Hemophilia and Hemophilia Federation of America to identify information‐educational needs and necessary capabilities for app design. A questionnaire based on these needs was then provided to physicians and patients for approval. In the second phase, the application prototype was designed according to the approved information‐educational needs and necessary capabilities and evaluated for usability using the standard Questionnaire for User Interface Satisfaction with 17 patients. Data were analysed using SPSS 20. Results: Sixty‐two information‐educational needs and capabilities with scores above 50% were included in the design. The designed application can provide the conditions for recording demographic information, medical history, treatment information, bleeding information, and injuries and accidents and provide the education on self‐management for patients. Moreover, it has the capabilities to set various reminders (taking medication, rehabilitation‐physical activities, and appointments with the therapist), set the treatment plan, send reports to the therapist, communicate with other persons with haemophilia and therapists, and record notes. Patients evaluated the application's usability at a “good” level. Conclusion: The haemophilia self‐management app can enhance patients’ quality of life by addressing their educational needs and supporting self‐management.
- Open Access
- 19 December 2025
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Predicting and classifying type 2 diabetes using a transparent ensemble model combining random forest, k-nearest neighbor, and neural networks
Diabetes is one of the major health challenges in today’s world, since chronic elevation of blood sugar can cause serious and sometimes irreparable damage to organs such as the heart, kidneys, and nervous system. Early detection of this disease plays a vital role in reducing its complications. However, machine learning and deep learning models often face distrust in medical settings due to their opaque, “black-box” nature. The aim of this study was to combine three machine learning algorithms using stacking and voting methods to propose a model for type 2 diabetes detection, and to increase transparency by using the explainability techniques LIME and SHAP to identify important features. This study used medical data from 768 Pima Indians Diabetes samples, including 8 features such as age, BMI, glucose, insulin, blood pressure, skin thickness, pregnancies, and family history. Data preprocessing included mean imputation for missing or zero values, Min–Max normalization, and classification into “Normal”, “Prediabetes”, and “Diabetes” based on fasting glucose thresholds. Feature selection was performed using Spearman correlation to retain the most relevant variables. A hybrid machine learning model was developed using three base models Neural Network (NN), k-Nearest Neighbors (KNN), and Random Forest (RF) with automated hyperparameter tuning. The outputs of these models were combined via stacking using a logistic regression (LR) meta-model and in parallel using a soft voting method. Nested cross-validation (5 outer and 5 inner folds) was applied to prevent data leakage and ensure robust evaluation. Model interpretability was assessed using LIME for local explanations and SHAP for global feature importance. Decision thresholds and influential feature regions were identified, and model calibration and decision curves evaluated clinical reliability. Models performance was evaluated using accuracy, precision, recall, specificity, F1-score, AUROC, Brier Score (1–B), and Expected Calibration Error (1–E). Statistical reliability was assessed using bootstrap resampling to compute 95% confidence intervals, as well as paired tests to compare the hybrid model with the base models and voting ensemble. Based on the evaluation metrics, the stacking ensemble achieved perfect performance for Class 0, with 100% accuracy, precision, recall, specificity, F1 score, and AUROC, alongside the highest calibration metrics (Brier Score: 99.9, ECE: 98.7). The Random Forest model also excelled, achieving 100% accuracy, precision, recall, specificity, and F1 score for Class 0 and Class 2. In contrast, the KNN model consistently underperformed, particularly for Class 0 (F1: 83.3, Precision: 83.3, Recall: 83.3). The Neural Network demonstrated strong recall for Class 0 (100%), while the voting ensemble showed balanced results but was slightly outperformed by the top ensemble methods. Explainable AI analyses using LIME and SHAP revealed that glucose was the most influential predictor for identifying the Pre-diabetes state. Both methods consistently identified a decision band between 0.35 and 0.47 (corresponding to 100–125 mg/dL) as the transition zone between “Normal” and “Prediabetes”, confirming the model’s alignment with WHO/ADA diagnostic criteria. The stacking model achieved perfect performance and superior calibration, outperforming all other models in type 2 diabetes prediction and classification. Explainability techniques (LIME and SHAP) identified glucose level, body mass index, and blood pressure as key predictive factors. This approach provides an accurate and interpretable tool for clinical decision support in healthcare systems.
- 19 December 2025
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessPredicting and classifying type 2 diabetes using a transparent ensemble model combining random forest, k-nearest neighbor, and neural networks
Diabetes is one of the major health challenges in today’s world, since chronic elevation of blood sugar can cause serious and sometimes irreparable damage to organs such as the heart, kidneys, and nervous system. Early detection of this disease plays a vital role in reducing its complications. However, machine learning and deep learning models often face distrust in medical settings due to their opaque, “black-box” nature. The aim of this study was to combine three machine learning algorithms using stacking and voting methods to propose a model for type 2 diabetes detection, and to increase transparency by using the explainability techniques LIME and SHAP to identify important features. This study used medical data from 768 Pima Indians Diabetes samples, including 8 features such as age, BMI, glucose, insulin, blood pressure, skin thickness, pregnancies, and family history. Data preprocessing included mean imputation for missing or zero values, Min–Max normalization, and classification into “Normal”, “Prediabetes”, and “Diabetes” based on fasting glucose thresholds. Feature selection was performed using Spearman correlation to retain the most relevant variables. A hybrid machine learning model was developed using three base models Neural Network (NN), k-Nearest Neighbors (KNN), and Random Forest (RF) with automated hyperparameter tuning. The outputs of these models were combined via stacking using a logistic regression (LR) meta-model and in parallel using a soft voting method. Nested cross-validation (5 outer and 5 inner folds) was applied to prevent data leakage and ensure robust evaluation. Model interpretability was assessed using LIME for local explanations and SHAP for global feature importance. Decision thresholds and influential feature regions were identified, and model calibration and decision curves evaluated clinical reliability. Models performance was evaluated using accuracy, precision, recall, specificity, F1-score, AUROC, Brier Score (1–B), and Expected Calibration Error (1–E). Statistical reliability was assessed using bootstrap resampling to compute 95% confidence intervals, as well as paired tests to compare the hybrid model with the base models and voting ensemble. Based on the evaluation metrics, the stacking ensemble achieved perfect performance for Class 0, with 100% accuracy, precision, recall, specificity, F1 score, and AUROC, alongside the highest calibration metrics (Brier Score: 99.9, ECE: 98.7). The Random Forest model also excelled, achieving 100% accuracy, precision, recall, specificity, and F1 score for Class 0 and Class 2. In contrast, the KNN model consistently underperformed, particularly for Class 0 (F1: 83.3, Precision: 83.3, Recall: 83.3). The Neural Network demonstrated strong recall for Class 0 (100%), while the voting ensemble showed balanced results but was slightly outperformed by the top ensemble methods. Explainable AI analyses using LIME and SHAP revealed that glucose was the most influential predictor for identifying the Pre-diabetes state. Both methods consistently identified a decision band between 0.35 and 0.47 (corresponding to 100–125 mg/dL) as the transition zone between “Normal” and “Prediabetes”, confirming the model’s alignment with WHO/ADA diagnostic criteria. The stacking model achieved perfect performance and superior calibration, outperforming all other models in type 2 diabetes prediction and classification. Explainability techniques (LIME and SHAP) identified glucose level, body mass index, and blood pressure as key predictive factors. This approach provides an accurate and interpretable tool for clinical decision support in healthcare systems.
- Open Access
- 11 November 2025
- Published by Springer Nature in Virtual Reality
- Vol. 29 (4) , 1-8
Examining the effects of virtual reality on anxiety and stress reduction in patients undergoing hemodialysis: a pre-post interventional study
- 11 November 2025
- Published by Springer Nature in Virtual Reality
- Vol. 29 (4) , 1-8
Open AccessExamining the effects of virtual reality on anxiety and stress reduction in patients undergoing hemodialysis: a pre-post interventional study