Muhammed Kürşad Uçar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Muhammed Kürşad Uçar has a total of 66 co-authors. Here are the top co-authors by number of shared publications:
Mehmet Recep Bozkurt
9 shared publications- Mehmet Recep Bozkurt
Electrical and Electronics Engineering Department, Sakarya University, Sakarya, Trkiye
7 shared publications Cahit Bilgin
6 shared publications- Cahit Bilgin
Sakarya University
5 shared publications - Derya Kandaz
Sakarya Üniversitesi
5 shared publications Uçar
4 shared publications
Latest Publications
- Open Access
- 2 January 2026
- Published by Springer Nature in BioMedical Engineering OnLine
- Vol. 25 (1)
A PRISMA-based systematic review on advances in identity recognition and authentication using human biometric signals (2018–2023)
This systematic review examines the effectiveness of physiological biometric signals in authentication and recognition systems by analyzing studies published between 2018 and 2023. Specifically, different biometric modalities (e.g., ECG, EEG, and PPG), commonly used datasets, signal processing techniques, and classification approaches are evaluated to assess their reported reliability and performance. In addition, the performance of multimodal biometric systems is compared with that of unimodal approaches. The review was conducted in accordance with the PRISMA 2020 guidelines. Relevant studies published between 2018 and 2023 were systematically retrieved from major databases, including EBSCO, PubMed, IEEE Xplore, Scopus, and Web of Science. A total of 2,064 records were initially identified, and after duplicate removal and eligibility screening, 80 articles were included in the final review. The study selection process is summarized using a PRISMA flow diagram. The reviewed studies indicate that ECG-based authentication systems report high average accuracy (98.6%), while multimodal biometric systems generally achieve accuracy levels exceeding 99%. Across modalities, deep learning–based approaches tend to outperform traditional machine learning methods. Dataset size and the choice of signal processing techniques were also found to influence reported performance outcomes. Overall, the findings suggest that biometric signal–based authentication systems demonstrate strong performance under the evaluation conditions reported in the literature. Multimodal fusion and deep learning approaches appear particularly promising, although reported results vary across datasets and protocols. Future research should prioritize larger and more diverse datasets, standardized evaluation benchmarks, and optimized signal processing pipelines to improve comparability and real-world applicability. Further studies on the integration of complementary biometric signals are also warranted.
- 2 January 2026
- Published by Springer Nature in BioMedical Engineering OnLine
- Vol. 25 (1)
Open AccessA PRISMA-based systematic review on advances in identity recognition and authentication using human biometric signals (2018–2023)
This systematic review examines the effectiveness of physiological biometric signals in authentication and recognition systems by analyzing studies published between 2018 and 2023. Specifically, different biometric modalities (e.g., ECG, EEG, and PPG), commonly used datasets, signal processing techniques, and classification approaches are evaluated to assess their reported reliability and performance. In addition, the performance of multimodal biometric systems is compared with that of unimodal approaches. The review was conducted in accordance with the PRISMA 2020 guidelines. Relevant studies published between 2018 and 2023 were systematically retrieved from major databases, including EBSCO, PubMed, IEEE Xplore, Scopus, and Web of Science. A total of 2,064 records were initially identified, and after duplicate removal and eligibility screening, 80 articles were included in the final review. The study selection process is summarized using a PRISMA flow diagram. The reviewed studies indicate that ECG-based authentication systems report high average accuracy (98.6%), while multimodal biometric systems generally achieve accuracy levels exceeding 99%. Across modalities, deep learning–based approaches tend to outperform traditional machine learning methods. Dataset size and the choice of signal processing techniques were also found to influence reported performance outcomes. Overall, the findings suggest that biometric signal–based authentication systems demonstrate strong performance under the evaluation conditions reported in the literature. Multimodal fusion and deep learning approaches appear particularly promising, although reported results vary across datasets and protocols. Future research should prioritize larger and more diverse datasets, standardized evaluation benchmarks, and optimized signal processing pipelines to improve comparability and real-world applicability. Further studies on the integration of complementary biometric signals are also warranted.
- Open Access
- 10 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 214033-214052
Continuous Blood Pressure Estimation Based on an Electrocardiography Signal Model
Objective: Accurate and continuous monitoring of arterial blood pressure (BP) is essential for effective management of cardiac and hypertension-related conditions. The limitations of non-invasive measurement techniques and the insufficient accuracy of invasive methods highlight the need for novel approaches. This study introduces a machine learning-based methodology for precise and continuous BP estimation by analyzing electrocardiography (ECG) signals in sub-frequency bands. Methodology: ECG and BP recordings from five subjects, sourced from the IEEE database, were utilized. The signals were segmented into specific sub-frequency bands (e.g., 2-40 Hz, 5-15 Hz), resulting in 594 machine learning models. A 10-second epoching process was applied to the ECG signals to enhance prediction accuracy. Statistical features were extracted from the time domain, and correlation analysis was performed to identify BP-relevant features for model training. Gaussian process regression, decision tree, and ensemble bagged tree algorithms were employed to develop and evaluate the prediction models. Results: Among the tested models, the best performance was achieved using the 2-40 Hz frequency band of the 10-second epoch ECG signals, yielding mean absolute percentage errors (MAPE) of 1.80 for systolic BP and 1.36 for diastolic BP, with a coefficient of determination (R2) of 0.99. Conclusion: This study demonstrates that analyzing ECG signals within specific frequency ranges strengthens their correlation with arterial blood pressure. Moreover, the inclusion of epoched ECG signals significantly enhances the predictive accuracy of BP estimation models, providing a promising foundation for non-invasive BP monitoring applications.
- 10 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 13, 214033-214052
Open AccessContinuous Blood Pressure Estimation Based on an Electrocardiography Signal Model
Objective: Accurate and continuous monitoring of arterial blood pressure (BP) is essential for effective management of cardiac and hypertension-related conditions. The limitations of non-invasive measurement techniques and the insufficient accuracy of invasive methods highlight the need for novel approaches. This study introduces a machine learning-based methodology for precise and continuous BP estimation by analyzing electrocardiography (ECG) signals in sub-frequency bands. Methodology: ECG and BP recordings from five subjects, sourced from the IEEE database, were utilized. The signals were segmented into specific sub-frequency bands (e.g., 2-40 Hz, 5-15 Hz), resulting in 594 machine learning models. A 10-second epoching process was applied to the ECG signals to enhance prediction accuracy. Statistical features were extracted from the time domain, and correlation analysis was performed to identify BP-relevant features for model training. Gaussian process regression, decision tree, and ensemble bagged tree algorithms were employed to develop and evaluate the prediction models. Results: Among the tested models, the best performance was achieved using the 2-40 Hz frequency band of the 10-second epoch ECG signals, yielding mean absolute percentage errors (MAPE) of 1.80 for systolic BP and 1.36 for diastolic BP, with a coefficient of determination (R2) of 0.99. Conclusion: This study demonstrates that analyzing ECG signals within specific frequency ranges strengthens their correlation with arterial blood pressure. Moreover, the inclusion of epoched ECG signals significantly enhances the predictive accuracy of BP estimation models, providing a promising foundation for non-invasive BP monitoring applications.
- 20 July 2025
- Published by Wiley in International Journal of Mental Health Nursing
- Vol. 34 (4) , e70084
Artificial Intelligence Based Approach to Self‐Sensitivity and Compassion Scores: Development of Prediction Models
This study investigates the usability of artificial intelligence and machine learning techniques to predict individuals' levels of self‐sensitivity and compassion. As self‐sensitivity and compassion significantly affect individuals' ability to cope with stress, this study aims to develop models to help effectively measure these psychosocial variables. The research covers Gaussian Process Regression (GPR), Neural Network Regression (Net), and Support Vector Machine (SVM) Regression models. The data were collected using the self‐sensitivity and compassion scales, and MAPE, MAE, SE, MSE, RMSE, R , and R 2 values were used as performance evaluation criteria for each model. The findings show that the GPR model provides highly accurate predictions for both scale types. The Net and SVM models also provided effective predictions, but GPR performed the best overall. Artificial intelligence and machine learning‐based models have emerged as practical tools for predicting self‐sensitivity and compassion scores. The GPR model is particularly notable for its high prediction accuracy. These findings offer important applications in nursing practice and the design of psychosocial interventions.
- 20 July 2025
- Published by Wiley in International Journal of Mental Health Nursing
- Vol. 34 (4) , e70084
Artificial Intelligence Based Approach to Self‐Sensitivity and Compassion Scores: Development of Prediction Models
This study investigates the usability of artificial intelligence and machine learning techniques to predict individuals' levels of self‐sensitivity and compassion. As self‐sensitivity and compassion significantly affect individuals' ability to cope with stress, this study aims to develop models to help effectively measure these psychosocial variables. The research covers Gaussian Process Regression (GPR), Neural Network Regression (Net), and Support Vector Machine (SVM) Regression models. The data were collected using the self‐sensitivity and compassion scales, and MAPE, MAE, SE, MSE, RMSE, R , and R 2 values were used as performance evaluation criteria for each model. The findings show that the GPR model provides highly accurate predictions for both scale types. The Net and SVM models also provided effective predictions, but GPR performed the best overall. Artificial intelligence and machine learning‐based models have emerged as practical tools for predicting self‐sensitivity and compassion scores. The GPR model is particularly notable for its high prediction accuracy. These findings offer important applications in nursing practice and the design of psychosocial interventions.
- Open Access
- 1 January 2025
- Published by Elsevier in SSRN Electronic Journal
A New Diagnostic Method for Chronic Obstructive Pulmonary Disease using the Photoplethysmography Signal and Hybrid Artificial Intelligence
Background and Purpose: Chronic obstructive pulmonary disease (COPD), is a primary public health issue globally and in our country, which continues to increase due to poor awareness of the disease and lack of necessary preventive measures. COPD is the result of a blockage of the air sacs known as alveoli within the lungs; it is a persistent sickness that causes difficulty in breathing, cough, and shortness of breath. COPD is characterized by breathing signs and symptoms and airflow challenge because of anomalies in the airways and alveoli that occurs as the result of significant exposure to harmful particles and gases. The spirometry test (breath measurement test), used for diagnosing COPD, is creating difficulties in reaching hospitals, especially in patients with disabilities or advanced disease and in children. To facilitate the diagnostic treatment and prevent these problems, it is far evaluated that using photoplethysmography (PPG) signal in the diagnosis of COPD disease would be beneficial in order to simplify and speed up the diagnosis process and make it more convenient for monitoring. A PPG signal includes numerous components, including volumetric changes in arterial blood that are related to heart activity, fluctuations in venous blood volume that modify the PPG signal, a direct current (DC) component that shows the optical properties of the tissues, and modest energy changes in the body. PPG has typically received the usage of a pulse oximeter, which illuminates the pores and skin and measures adjustments in mild absorption. PPG occurring with every heart rate is an easy signal to measure. PPG signal is modeled by machine learning to predict COPD. Methods: During the studies, the PPG signal was cleaned of noise, and a brand-new PPG signal having three low-frequency bands of the PPG was obtained. Each of the four signals extracted 25 features. An aggregate of 100 features have been extracted. Additionally, weight, height, and age were also used as characteristics. In the feature selection process, we employed the Fisher method. The intention of using this method is to improve performance. Results: This improved PPG prediction models have an accuracy rate of 0.95 performance value for all individuals. Classification algorithms used in feature selection algorithm has contributed to a performance increase. Conclusion: According to the findings, PPG-based COPD prediction models are suitable for usage in practice.
- 1 January 2025
- Published by Elsevier in SSRN Electronic Journal
Open AccessA New Diagnostic Method for Chronic Obstructive Pulmonary Disease using the Photoplethysmography Signal and Hybrid Artificial Intelligence
Background and Purpose: Chronic obstructive pulmonary disease (COPD), is a primary public health issue globally and in our country, which continues to increase due to poor awareness of the disease and lack of necessary preventive measures. COPD is the result of a blockage of the air sacs known as alveoli within the lungs; it is a persistent sickness that causes difficulty in breathing, cough, and shortness of breath. COPD is characterized by breathing signs and symptoms and airflow challenge because of anomalies in the airways and alveoli that occurs as the result of significant exposure to harmful particles and gases. The spirometry test (breath measurement test), used for diagnosing COPD, is creating difficulties in reaching hospitals, especially in patients with disabilities or advanced disease and in children. To facilitate the diagnostic treatment and prevent these problems, it is far evaluated that using photoplethysmography (PPG) signal in the diagnosis of COPD disease would be beneficial in order to simplify and speed up the diagnosis process and make it more convenient for monitoring. A PPG signal includes numerous components, including volumetric changes in arterial blood that are related to heart activity, fluctuations in venous blood volume that modify the PPG signal, a direct current (DC) component that shows the optical properties of the tissues, and modest energy changes in the body. PPG has typically received the usage of a pulse oximeter, which illuminates the pores and skin and measures adjustments in mild absorption. PPG occurring with every heart rate is an easy signal to measure. PPG signal is modeled by machine learning to predict COPD. Methods: During the studies, the PPG signal was cleaned of noise, and a brand-new PPG signal having three low-frequency bands of the PPG was obtained. Each of the four signals extracted 25 features. An aggregate of 100 features have been extracted. Additionally, weight, height, and age were also used as characteristics. In the feature selection process, we employed the Fisher method. The intention of using this method is to improve performance. Results: This improved PPG prediction models have an accuracy rate of 0.95 performance value for all individuals. Classification algorithms used in feature selection algorithm has contributed to a performance increase. Conclusion: According to the findings, PPG-based COPD prediction models are suitable for usage in practice.
- Open Access
- 30 December 2024
- Published by Balikesir Medical Journal in Balıkesir Medical Journal
- Vol. 8 (3) , 156-162
Contribution of Image Processing in Chest X-Ray to Early Diagnosis in Radiological Evaluation of Lung Cancer
This paper attempts to investigate the importance of image processing by chest X-rays (CXRs) in the early diagnosis of lung cancer. It analyzes the contributions of CXRs to the radiological assessment of lung cancer, discussing their benefits and flaws and proposing image processing methods for improving their performance. The research evaluates the performance of CXR versus other imaging methods including CT and focuses on the early diagnosis which is crucial for enhancing patients’ outcome Moreover, the paper delves into the latest progress in image technology and shows how it is used in improving the accuracy of chest radiographs in the diagnosis of lung cancer.
- 30 December 2024
- Published by Balikesir Medical Journal in Balıkesir Medical Journal
- Vol. 8 (3) , 156-162
Open AccessContribution of Image Processing in Chest X-Ray to Early Diagnosis in Radiological Evaluation of Lung Cancer
This paper attempts to investigate the importance of image processing by chest X-rays (CXRs) in the early diagnosis of lung cancer. It analyzes the contributions of CXRs to the radiological assessment of lung cancer, discussing their benefits and flaws and proposing image processing methods for improving their performance. The research evaluates the performance of CXR versus other imaging methods including CT and focuses on the early diagnosis which is crucial for enhancing patients’ outcome Moreover, the paper delves into the latest progress in image technology and shows how it is used in improving the accuracy of chest radiographs in the diagnosis of lung cancer.
- RetractionOpen Access
- 29 July 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (25) , 15945
Retraction Note: Automatic detection of respiratory arrests in OSA patients using PPG and machine learning techniques
- 29 July 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (25) , 15945
RetractionOpen AccessRetraction Note: Automatic detection of respiratory arrests in OSA patients using PPG and machine learning techniques
- RetractionOpen Access
- 16 July 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (24) , 15191
Retraction Note: Automatic sleep staging in obstructive sleep apnea patients using photoplethysmography, heart rate variability signal and machine learning techniques
- 16 July 2024
- Published by Springer Nature in Neural Computing and Applications
- Vol. 36 (24) , 15191
RetractionOpen AccessRetraction Note: Automatic sleep staging in obstructive sleep apnea patients using photoplethysmography, heart rate variability signal and machine learning techniques
- 6 April 2024
- Published by Springer Nature
- p. 222-238
Machine Learning Based Detection of Forest Fires Using Sensors Placed in Forests
- 6 April 2024
- Published by Springer Nature
- p. 222-238
Machine Learning Based Detection of Forest Fires Using Sensors Placed in Forests
- Open Access
- 23 December 2023
- Published by Turkish Journal of Forecasting in Turkish Journal of Forecasting
- Vol. 07 (2) , 17-22
Hybrid AI-based Voice Authentication
Biometric authentication systems reveal individuals' physical or behavioral uniqueness and identify them by comparing them with existing records. Today, many biometric recognition systems, such as fingerprint reading, palm reading, and face reading, are being studied and used. The human voice is also among the techniques used for this purpose. Due to this feature, the human voice performs secure transactions and authentication in various fields. Based on these voice features, we used a dataset of 66,569 voice recordings. The voice recordings were revised to include six sentences of at least six words each from 24 different people to get the maximum benefit from the dataset. The voices in the reduced dataset were labeled as sentences belonging to the same person and sentences belonging to different people and converted into matrix form. A biometric recognition study resulted in a correlation score of 0.88. As a result of these processes, the feasibility of a voice biometric recognition system with artificial intelligence has been demonstrated.
- 23 December 2023
- Published by Turkish Journal of Forecasting in Turkish Journal of Forecasting
- Vol. 07 (2) , 17-22
Open AccessHybrid AI-based Voice Authentication
Biometric authentication systems reveal individuals' physical or behavioral uniqueness and identify them by comparing them with existing records. Today, many biometric recognition systems, such as fingerprint reading, palm reading, and face reading, are being studied and used. The human voice is also among the techniques used for this purpose. Due to this feature, the human voice performs secure transactions and authentication in various fields. Based on these voice features, we used a dataset of 66,569 voice recordings. The voice recordings were revised to include six sentences of at least six words each from 24 different people to get the maximum benefit from the dataset. The voices in the reduced dataset were labeled as sentences belonging to the same person and sentences belonging to different people and converted into matrix form. A biometric recognition study resulted in a correlation score of 0.88. As a result of these processes, the feasibility of a voice biometric recognition system with artificial intelligence has been demonstrated.