Volkan Türkmen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Volkan Türkmen has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Sena Dilşad Akçakaya
Department of Nutrition and Dietetics, Gülhane Health Sciences Institute, University of Health Sciences, Ankara, Turkey
3 shared publications - Özge Mengi Çelik
Department of Nutrition and Dietetics, Gulhane Faculty of Health Sciences, University of Health Sciences, Ankara, Türkiye
3 shared publications - Aslı Hilal Güzelalp
Department of Nutrition and Dietetics, Faculty of Health Sciences Istanbul Rumeli University Istanbul Türkiye
1 shared publication - Berna Betül Cihan
Department of Nutrition and Dietetics, Gülhane Health Sciences Institute, University of Health Sciences, Ankara, Turkey
1 shared publication - Dilek Tezcan
Department of Rheumatology, Gülhane Training and Research Hospital, Ankara, Turkiye
1 shared publication Dogukan Ozen
Saglik Bilimleri Universitesi Gulhane Saglik Bilimleri Enstitusu
1 shared publication
Latest Publications
- Open Access
- 20 October 2026
- Published by Springer Nature in BMC Psychology
Cross-cultural adaptation and psychometric validation of the Turkish intuitive eating scale-3: a methodological study
- 20 October 2026
- Published by Springer Nature in BMC Psychology
Open AccessCross-cultural adaptation and psychometric validation of the Turkish intuitive eating scale-3: a methodological study
- Open Access
- 8 October 2026
- Published by Wiley
Comparative Performance of Machine Learning Algorithms for Multiclass Classification of Earnings Status in Thoroughbred Racehorses in Türkiye
Background: Most previous studies relied on traditional regression models to predict Thoroughbred earnings, highlighting the need for machine learning approaches using lifetime earnings data to improve prediction accuracy and economic insight. Objectives: This study compared the performance of selected machine learning algorithms for the multiclass classification of cumulative earnings status in Thoroughbred racehorses in Türkiye. Study design: This retrospective registry-based study was conducted using publicly available records maintained by the Turkish Jockey Club (TJK). Each horse was considered as the unit of analysis. Methods: Using Turkish Jockey Club data, 14,233 Thoroughbred racehorses were categorized by their median cumulative earnings into three distinct groups: non-earners, low earners, and high earners. Four machine learning algorithms (k-Nearest Neighbors, C5.0, Random Forest and Gradient Boosting Machine) were evaluated using three different training-test splits and 10-fold cross-validation (10 repetitions) methods. Models were trained on both original and Random Oversampling-balanced training sets and evaluated on the original test set using accuracy, macro-averaged metrics, G-mean, multiclass Matthews correlation coefficient, and multiclass AUC. Results: The GBM model achieved the highest accuracy under the 75:25 split with the original training data, although gains over the majority-class baseline were limited. Random Oversampling improved class-balanced performance despite reducing overall accuracy. Variable-importance analyses identified number of races, age at first race, and racecourse type as the most influential predictors of lifetime earnings status. Main limitations: Several other important factors known to influence race performance, such as injury history, training conditions, and the complete pedigree structure could not be incorporated into the present study. Conclusion: These findings should be interpreted as model-based associations derived from retrospective registry data rather than as causal effects. Overall, the results suggest that machine learning algorithms can be used to benchmark multiclass classification performance for earnings-status categories in Thoroughbred racehorses, while highlighting the importance of evaluating class imbalance and class-specific predictive performance.
- 8 October 2026
- Published by Wiley
Open AccessComparative Performance of Machine Learning Algorithms for Multiclass Classification of Earnings Status in Thoroughbred Racehorses in Türkiye
Background: Most previous studies relied on traditional regression models to predict Thoroughbred earnings, highlighting the need for machine learning approaches using lifetime earnings data to improve prediction accuracy and economic insight. Objectives: This study compared the performance of selected machine learning algorithms for the multiclass classification of cumulative earnings status in Thoroughbred racehorses in Türkiye. Study design: This retrospective registry-based study was conducted using publicly available records maintained by the Turkish Jockey Club (TJK). Each horse was considered as the unit of analysis. Methods: Using Turkish Jockey Club data, 14,233 Thoroughbred racehorses were categorized by their median cumulative earnings into three distinct groups: non-earners, low earners, and high earners. Four machine learning algorithms (k-Nearest Neighbors, C5.0, Random Forest and Gradient Boosting Machine) were evaluated using three different training-test splits and 10-fold cross-validation (10 repetitions) methods. Models were trained on both original and Random Oversampling-balanced training sets and evaluated on the original test set using accuracy, macro-averaged metrics, G-mean, multiclass Matthews correlation coefficient, and multiclass AUC. Results: The GBM model achieved the highest accuracy under the 75:25 split with the original training data, although gains over the majority-class baseline were limited. Random Oversampling improved class-balanced performance despite reducing overall accuracy. Variable-importance analyses identified number of races, age at first race, and racecourse type as the most influential predictors of lifetime earnings status. Main limitations: Several other important factors known to influence race performance, such as injury history, training conditions, and the complete pedigree structure could not be incorporated into the present study. Conclusion: These findings should be interpreted as model-based associations derived from retrospective registry data rather than as causal effects. Overall, the results suggest that machine learning algorithms can be used to benchmark multiclass classification performance for earnings-status categories in Thoroughbred racehorses, while highlighting the importance of evaluating class imbalance and class-specific predictive performance.
- Open Access
- 21 September 2026
- Published by Springer Nature in Journal of Eating Disorders
- Vol. 14 (1)
Validation and reliability of the Turkish Binge Eating Disorder Scale: a community-based study
- 21 September 2026
- Published by Springer Nature in Journal of Eating Disorders
- Vol. 14 (1)
Open AccessValidation and reliability of the Turkish Binge Eating Disorder Scale: a community-based study
- 4 August 2026
- Published by Springer Nature in Bratislava Medical Journal
- Vol. 127 (11) , 5263-5274
Comparative Analysis of Novel Inflammatory Indices for Predicting Relapse in Granulomatous Mastitis: Lymphocyte-to-Monocyte Ratio Emerges as the Superior Biomarker
- 4 August 2026
- Published by Springer Nature in Bratislava Medical Journal
- Vol. 127 (11) , 5263-5274
Comparative Analysis of Novel Inflammatory Indices for Predicting Relapse in Granulomatous Mastitis: Lymphocyte-to-Monocyte Ratio Emerges as the Superior Biomarker
- Open Access
- 4 July 2026
- Published by Springer Nature in BMC Geriatrics
- Vol. 26 (1)
Cross-cultural adaptation and validation of the mindful eating behavior scale in community-dwelling older adults
Mindful eating is a fundamental approach for promoting healthy dietary behaviors and well-being in older age. However, there is currently no validated scale available to assess mindful eating specifically among older adults in the Turkish population. Therefore, a reliable instrument is needed to address this gap and enable accurate assessment of eating awareness in this demographic. The primary aim of this study was to evaluate the validity and reliability of the Turkish version of the Mindful Eating Behavior Scale (MEBS-TR) in adults aged 65 years and older. This methodological study included 424 individuals aged 65 and over (61.3% female) with a mean age of 70.87±6.15 years. Data were collected using a Personal Information Form, the MEBS, and the Mediterranean Diet Adherence Screener (MEDAS). The construct validity of the scale was assessed using Minimum Rank Factor Analysis (MRFA) with a polychoric correlation matrix and Confirmatory Factor Analysis (CFA) with Mean and Variance Adjusted Maximum Likelihood (MLMV) estimation. Concurrent validity was examined via Spearman's correlation between MEBS-TR and MEDAS scores. For reliability analysis, Cronbach's alpha (α) and McDonald's omega (ω) internal consistency coefficients were calculated, along with test-retest reliability using the Intraclass Correlation Coefficient (ICC). The MRFA results confirmed the original 17-item, four-factor structure (Focused Eating, Hunger and Satiety Cues, Eating with Awareness, and Eating without Distraction), which explained 81.81% of the total variance. CFA demonstrated that this four-factor model had acceptable model fit indices (χ2/df = 2.777, CFI = 0.941, TLI = 0.928, RMSEA = 0.065). The internal consistency coefficients for the total scale were found to be high (Cronbach's α = 0.861, McDonald's ω = 0.867). Test-retest reliability, assessed with 150 participants, was found to be excellent (ICC = 0.969). A positive and statistically significant relationship was found between the MEBS-TR total score and the MEDAS total score (rho = .163, p = .001). These findings demonstrate that the MEBS-TR is a valid and reliable instrument for assessing mindful eating behaviors among adults aged 65 years and older in Türkiye.
- 4 July 2026
- Published by Springer Nature in BMC Geriatrics
- Vol. 26 (1)
Open AccessCross-cultural adaptation and validation of the mindful eating behavior scale in community-dwelling older adults
Mindful eating is a fundamental approach for promoting healthy dietary behaviors and well-being in older age. However, there is currently no validated scale available to assess mindful eating specifically among older adults in the Turkish population. Therefore, a reliable instrument is needed to address this gap and enable accurate assessment of eating awareness in this demographic. The primary aim of this study was to evaluate the validity and reliability of the Turkish version of the Mindful Eating Behavior Scale (MEBS-TR) in adults aged 65 years and older. This methodological study included 424 individuals aged 65 and over (61.3% female) with a mean age of 70.87±6.15 years. Data were collected using a Personal Information Form, the MEBS, and the Mediterranean Diet Adherence Screener (MEDAS). The construct validity of the scale was assessed using Minimum Rank Factor Analysis (MRFA) with a polychoric correlation matrix and Confirmatory Factor Analysis (CFA) with Mean and Variance Adjusted Maximum Likelihood (MLMV) estimation. Concurrent validity was examined via Spearman's correlation between MEBS-TR and MEDAS scores. For reliability analysis, Cronbach's alpha (α) and McDonald's omega (ω) internal consistency coefficients were calculated, along with test-retest reliability using the Intraclass Correlation Coefficient (ICC). The MRFA results confirmed the original 17-item, four-factor structure (Focused Eating, Hunger and Satiety Cues, Eating with Awareness, and Eating without Distraction), which explained 81.81% of the total variance. CFA demonstrated that this four-factor model had acceptable model fit indices (χ2/df = 2.777, CFI = 0.941, TLI = 0.928, RMSEA = 0.065). The internal consistency coefficients for the total scale were found to be high (Cronbach's α = 0.861, McDonald's ω = 0.867). Test-retest reliability, assessed with 150 participants, was found to be excellent (ICC = 0.969). A positive and statistically significant relationship was found between the MEBS-TR total score and the MEDAS total score (rho = .163, p = .001). These findings demonstrate that the MEBS-TR is a valid and reliable instrument for assessing mindful eating behaviors among adults aged 65 years and older in Türkiye.
- 8 April 2024
- Published by Oxford University Press (OUP) in Military Medicine
- Vol. 189 (9-10) , e2200-e2205
Evaluation of Telemedicine Support for Medics from the Perspective of Course Instructors
Introduction: Health services in the theater/district areas often contain difficulties. Although telemedicine has a huge potential to support medics in the area, there are challenges as well. Our aim is to evaluate the telemedicine support that can be provided to the medics from the perspective of course instructors. Materials and Methods: Our study was carried out in The University of Health Sciences, Türkiye. All of the instructors of a medic course were asked to fill a 13-question questionnaire and Likert-type scale, sent via e-mail. Among 79 instructors, 71 of them (55 physicians, 16 nonphysicians) responded and were included in the study. The distributions of the categorical variables in the groups were analyzed with the Pearson Chi-square. Results: Of the 71 participants, 37 (52.11%) stated that there is a need for legal and ethical regulations for telemedicine (medical liability, malpractice, obtaining consent from the casualties, civilians, cross border missions, and rank-related problems) and patients’ rights, additionally, to clarify the responsibilities of the doctors giving telemedicine support and the medics in the area. It was observed that physicians’ and nonphysician group’s opinions about the telepsychotherapy were statistically significantly different (χ2 = 8.675, P = .013). Conclusions: Most of the instructors believed that telemedicine could carry the knowledge and skills of specialist physicians to the field of operation. Access to high-quality health services in a short time through telemedicine support can increase the courage and commitment of the personnel. It is thought that with telemedicine, medics will not feel alone in the field, their worries about making wrong decisions will decrease, and their knowledge and skills will increase. Participants were particularly concerned about who will be held responsible for problems arising from interventions carried out using telemedicine, and strict legal and medical regulations are needed.
- 8 April 2024
- Published by Oxford University Press (OUP) in Military Medicine
- Vol. 189 (9-10) , e2200-e2205
Evaluation of Telemedicine Support for Medics from the Perspective of Course Instructors
Introduction: Health services in the theater/district areas often contain difficulties. Although telemedicine has a huge potential to support medics in the area, there are challenges as well. Our aim is to evaluate the telemedicine support that can be provided to the medics from the perspective of course instructors. Materials and Methods: Our study was carried out in The University of Health Sciences, Türkiye. All of the instructors of a medic course were asked to fill a 13-question questionnaire and Likert-type scale, sent via e-mail. Among 79 instructors, 71 of them (55 physicians, 16 nonphysicians) responded and were included in the study. The distributions of the categorical variables in the groups were analyzed with the Pearson Chi-square. Results: Of the 71 participants, 37 (52.11%) stated that there is a need for legal and ethical regulations for telemedicine (medical liability, malpractice, obtaining consent from the casualties, civilians, cross border missions, and rank-related problems) and patients’ rights, additionally, to clarify the responsibilities of the doctors giving telemedicine support and the medics in the area. It was observed that physicians’ and nonphysician group’s opinions about the telepsychotherapy were statistically significantly different (χ2 = 8.675, P = .013). Conclusions: Most of the instructors believed that telemedicine could carry the knowledge and skills of specialist physicians to the field of operation. Access to high-quality health services in a short time through telemedicine support can increase the courage and commitment of the personnel. It is thought that with telemedicine, medics will not feel alone in the field, their worries about making wrong decisions will decrease, and their knowledge and skills will increase. Participants were particularly concerned about who will be held responsible for problems arising from interventions carried out using telemedicine, and strict legal and medical regulations are needed.
- Open Access
- 30 December 2023
- Published by Veterinary Pharmacology and Toxicology Association in Veteriner Farmakoloji ve Toksikoloji Derneği Bülteni
- Vol. 14 (3) , 118-132
APPLICATIONS of REGRESSION ALGORITHMS in MACHINE LEARNING in VETERINARY MEDICINE
Hızla artan veriler, özellikle giyilebilir teknolojiler, sensörler ve internet bağlantılı akıllı ürünler (IoT) gibi yeni veri kaynaklarından akan daha büyük, daha karmaşık veri kümeleri makine öğrenmesi algoritmalarına olan ihtiyacı her geçen gün arttırmaktadır. Geleneksel veri işleme yazılımlarının başa çıkamadığı büyük hacimli veri kümeleri her alanda olduğu gibi veteriner hekimlik alanında da yeni fırsatlar sunmakta ve daha önce üstesinden gelinemeyen sorunlar için yeni çözüm yolları üretebilme potansiyeline sahip olduğu görülmektedir. Bu derleme çalışmasında literatürde sıkça karşılaşılan makine öğrenmesi regresyon algoritmaları tanıtılmış ve veteriner hekimliği alanında uygulamalarına ilişkin kısa örnekler verilmiştir. Bunun yanında bu algoritmaların veteriner hekimliği alanındaki potansiyeline kısaca değinilmiştir.
- 30 December 2023
- Published by Veterinary Pharmacology and Toxicology Association in Veteriner Farmakoloji ve Toksikoloji Derneği Bülteni
- Vol. 14 (3) , 118-132
Open AccessAPPLICATIONS of REGRESSION ALGORITHMS in MACHINE LEARNING in VETERINARY MEDICINE
Hızla artan veriler, özellikle giyilebilir teknolojiler, sensörler ve internet bağlantılı akıllı ürünler (IoT) gibi yeni veri kaynaklarından akan daha büyük, daha karmaşık veri kümeleri makine öğrenmesi algoritmalarına olan ihtiyacı her geçen gün arttırmaktadır. Geleneksel veri işleme yazılımlarının başa çıkamadığı büyük hacimli veri kümeleri her alanda olduğu gibi veteriner hekimlik alanında da yeni fırsatlar sunmakta ve daha önce üstesinden gelinemeyen sorunlar için yeni çözüm yolları üretebilme potansiyeline sahip olduğu görülmektedir. Bu derleme çalışmasında literatürde sıkça karşılaşılan makine öğrenmesi regresyon algoritmaları tanıtılmış ve veteriner hekimliği alanında uygulamalarına ilişkin kısa örnekler verilmiştir. Bunun yanında bu algoritmaların veteriner hekimliği alanındaki potansiyeline kısaca değinilmiştir.