EAErhan Akdoğan

Erhan Akdoğan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Erhan Akdoğan has a total of 126 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Fuzzy logic approaches for forecasting daily global solar radiation: A comparative analysis of climatic variables

    by Berke Oğulcan Parlak, Erhan Akdoğan

    Accurate prediction of solar radiation is important for the design and optimization of photo-voltaic systems and for efficient energy management and sustainability planning. However, in many regions direct solar radiation measurements are scarce or not available, which makes the forecasting from available climatic variables more and more important. This study addresses this challenge by developing three simple-structured fuzzy logic models to predict daily global solar radiation using three different combinations of commonly recorded weather inputs: temperature and relative humidity, temperature and wind speed, and relative humidity and wind speed. The models were trained with data from Ankara, Türkiye and evaluated using mean absolute percentage error and root mean square error metrics. Comparing the models tested, the relative humidity and wind speed model resulted in the best forecasting performance with a mean absolute percentage error of 9.07% and a root mean square error of 0.85 kWh/m²-day. It also outperformed the feedforward artificial neural network model used for comparison, which produced an error of 14.38%. Results indicate that wind speed has more direct effects on estimation of solar radiation while the effect of temperature appears more indirect. A major contribution of the present study is that reasonably simple fuzzy logic models can provide good prediction performance if the right input variables are selected. This is particularly important when compared to more complex artificial intelligence methods, which usually require larger data sets and a larger number of input parameters. In this sense, the proposed approach also helps to overcome limitations reported in previous studies, such as the use of high-dimensional input sets and models developed mainly for short-term forecasting. Overall, the results show that fuzzy logic can be an accurate and computationally efficient alternative for solar radiation forecasting, especially when the available data are limited.

  • MammosighTR: Nationwide Breast Cancer Screening Mammogram Dataset with BI-RADS Annotations for Artificial Intelligence Applications

    by Ural Koç, Muhammed Said Beşler, Ebru Akçapınar Sezer, Emrah Karakaş, Yaşar Alper Özkaya, Şehnaz Evrimler, Ahmet Yalçın, Alper Kızıloğlu, Uğur Kesimal, Meltem Oruç, et al.

    “Just Accepted” papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. The MammosighTR dataset, derived from Türkiye's national breast cancer screening mammography program, provides BI-RADS-labeled mammograms with detailed annotations on breast composition and lesion quadrant location, which may be useful for developing and testing AI models in breast cancer detection. ©RSNA, 2025


  • Open Access

    Development of artificial intelligence and multi-sensor-based dexterity assessment system: performance evaluation

    by Mehmet Emin Aktan, Sena Zeybek Kılıç, Erhan Akdoğan, Tuğçe Özekli Mısırlıoğlu, Deniz Palamar

    Manual dexterity tests are essential for diagnosing diseases and evaluating professional skills that require fine motor control. Traditional assessments often depend on expert supervision, leading to delays, subjectivity, and inaccuracies. This study introduces an automated dexterity assessment system that integrates multiple sensors and artificial intelligence algorithms for high-precision evaluation. The system classifies hand movements, analyzes muscle contraction levels, and determines hold-release durations using electromyography (EMG), inertial measurement units (IMU), and image processing techniques. An expert system interprets the multimodal sensor data and presents the results to clinicians. A performance evaluation with 20 participants demonstrated the system’s capability to assess hand dexterity automatically and accurately. Graphical Abstract

  • Investigation of Delays and Connection Stability in Teleoperation via VR Capabilities for Remote Exoskeleton-Aided Physiotherapy

    by Piotr Falkowski, Krzysztof Zawalski, Kajetan Jeznach, Jan Oleksiuk, Mehmet Emin Aktan, Vasfi Emre Ömürlü, Paweł Niedbalski, Erhan Akdoğan, Piotr Kołodziejski

    The research paper examines the challenges of remote control delays in exoskeleton-aided physiotherapy. It focuses on the implications of telerehabilitation requirements for safe treatment. As the global share of disabled and elderly people increases, the demand for effective teleoperated physiotherapy solutions rises, necessitating precise control over robotic systems. The study utilizes a mixed-methods approach to assess delays in communication between a rehabilitation exoskeleton and the control system's based on the digital twin in VR (virtual reality) components at different stages. Moreover, additional delays coming from distant operations are assessed. Key findings reveal that the average latency for joint control on shorter distances remains below 300 ms-considered acceptable for comfortable manual teleoperation. The delays between setting the device's configuration and reaching it by a physical exoskeleton differ for particular joints and depend mainly on the torque-to-moment of inertia ratio. The total communication cycle should not exceed 400 ms, which can be further reduced by using a 6G network in the future or ADS (Automation Device Specification) communication protocol instead of the database. Ultimately, the study provides a foundation for developing algorithms that can mitigate delays and improve the efficacy of telehealth solutions with physically supportive devices.


  • Test Automation for Deterministic Communication Networks Using Chained Orthogonal Arrays

    by Muhammed Efe Altuntaş, İbRahim Hökelek, M. Selim Demir, Muhammed Şeker, Pınar Savaş, Eyüp Can Akpolat, Sabri Barış Yavuztürk, Erhan Akdoğan

    ARINC 664 is one of the most popular Ethernet-based deterministic network (DTN) technologies for safety-critical avionics systems. A typical ARINC 664 network can support several thousands of Virtual Links (VLs), each of which has hundreds of thousands of combinations for its configuration. Orthogonal Arrays (OAs) method is a formalism to generate a test suite with a feasible size and a desired coverage of features for complex systems. Since the test space for DTN features is prohibitively large, a single application of OAs may inadvertently eliminate many combinations of two-way interactions. In this paper, we propose to develop a test automation system that uses multiple and consecutive applications of OAs to generate test configurations for DTN. The test factors representing DTN configuration parameters are divided into two groups, where the first group includes the number of VLs passing through each port while the second group includes the configuration parameters for VLs. Then, OAs are constructed sequentially for each group such that an output element of the first OA is used as one of the inputs for the second OA. We present a system architecture for a test automation system capable of verifying real-time network operations, where the outputs of the chained OAs are utilized to generate executable test scripts including DTN configurations and traffics. As a case study, a test automation system is successfully implemented for ARINC 664 switch.

  • Open Access

    Systematic review of mechanical designs of rehabilitation exoskeletons for lower-extremity

    by Krzysztof Zawalski, Jan Oleksiuk, Piotr Falkowski, Bazyli Leczkowski, Zbigniew Pilat, Mehmet Emin Aktan, Erhan Akdogan, Paweł Niedbalski, Maciej Mazowiec, Mateusz Wawrzyniak, et al.

    Exoskeletons of lower extremities are used mainly for gait treatment in physical rehabilitation. However, they are also capable of being involved in other types of exercises. Nevertheless, their structure needs to be adequately adjusted for such applications. To analyse approaches to that, this review paper investigates the mechanical designs of rehabilitation exoskeletons for lower extremities. The study seeks to identify best practices in designing and implementing these devices by analysing fifty-two articles. It covers aspects such as kinematic structures, materials used, types of drives, and the range of exercises. Standard design features include multiple degrees of freedom, primarily at the hip, knee, and ankle joints, and using lightweight materials to enhance mobility and reduce power consumption. The review also discusses the advantages of different driving systems. The findings provide valuable insights for developing effective and safe rehabilitation exoskeletons, contributing to improved patient outcomes in physiotherapy and rehabilitation settings.


  • DESIGN, DEVELOPMENT AND HYBRID IMPEDANCE CONTROL OF AN ANKLE REHABILITATION ROBOT

    by Emre Yildirim, Mehmet Emi̇n Aktan, Erhan Akdoğan, Tuğçe Özekli̇ Misirlioğlu, Deni̇z Palamar

    In this study, a hybrid impedance control-based portable ankle rehabilitation robot that can perform therapeutic exercises for the ankle has been developed. This 1-DOF robot can perform plantar flexion and dorsiflexion movements for the ankle. The capacity of the robotic system to perform therapeutic exercises was tested with subjects. Isotonic and resistive exercises were tested with 10 healthy subjects and stretching exercise was tested with a patient. The results showed that the hybrid impedance controlled robotic system can successfully perform passive stretching, active isotonic and active resistive exercises by providing a safe human-robot interaction. Especially when a sudden resistance increase occurs in the joint, the hybrid impedance controller acts like a physiotherapist and performs the movement without damaging the joint.


  • Skin Lesion Classification Using Convolutional Neural Network and ABCD Rule

    by Ezgi Kestek, Mehmet Emin Aktan, Erhan Akdoğan

    Skin cancer, which can occur in any part of the human skin, is one of the common and serious types of cancer. Accurate diagnosis and segmentation of lesions are crutial to the early diagnosis. Computer-aided diagnosis make important contributions to help doctors in the diagnosis of cancer from skin images. The most important factor for such systems to reveal the accurate results is the correct feature extraction. In this study, a model for the classification of seven types of skin lesions was developed by combining the features of CNN-based feature extraction and the ABCD rule, which is widely used in the clinic. The model was evaluated on HAM10000 well-known dataset. The classification results obtained with different combinations of features and machine learning algorithms were compared. According to the results, the best classification accuracy was obtained with the Cosine Similarity Classifier with 96.4% when the features determined by CNN and the features in the ABCD rule were used together.