Altınbaş Üniversitesi Kurumsal Akademik Arşivi
DSpace@Altınbaş, Altınbaş Üniversitesi tarafından doğrudan ve dolaylı olarak yayınlanan; kitap, makale, tez, bildiri, rapor, araştırma verisi gibi tüm akademik kaynakları uluslararası standartlarda dijital ortamda depolar, Üniversitenin akademik performansını izlemeye aracılık eder, kaynakları uzun süreli saklar ve telif haklarına uygun olarak Açık Erişime sunar.
Güncel Gönderiler
Enhancing Driving Control via Speech Recognition Utilizing Influential Parameters in Deep Learning Techniques
(Multidisciplinary Digital Publishing Institute (MDPI), 2025) Hussein, Hasan H.; Karan, Oğuz; Kurnaz, Sefer
This study investigates the enhancement of automated driving and command control through speech recognition using a Deep Neural Network (DNN). The method depends on some sequential stages such as noise removal, feature extraction from the audio file, and their classification using a neural network. In the proposed approach, the variables that affect the results in the hidden layers were extracted and stored in a vector to classify them and issue the most influential ones for feedback to the hidden layers in the neural network to increase the accuracy of the result. The result was 93% in terms of accuracy and with a very good response time of 0.75 s, with PSNR 78 dB. The proposed method is considered promising and is highly satisfactory to users. The results encouraged the use of more commands, more data processing, more future exploration, and the addition of sensors to increase the efficiency of the system and obtain more efficient and safe driving, which is the main goal of this research.
Automated Age Estimation from OPG Images and Patient Records Using Deep Feature Extraction and Modified Genetic-Random Forest
(2025) Uçan, Gülfem Özlü; Gwassi, Omar Abboosh Hussein; Apaydın, Burak Kerem; Uçan, Bahadır
Background/Objectives: Dental age estimation is a vital component of forensic science, helping to determine the identity and actual age of an individual. However, its effectiveness is challenged by methodological variability and biological differences between individuals. Therefore, to overcome the drawbacks such as the dependence on manual measurements, requiring a lot of time and effort, and the difficulty of routine clinical application due to large sample sizes, we aimed to automatically estimate tooth age from panoramic radiographs (OPGs) using artificial intelligence (AI) algorithms. Methods: Two-Dimensional Deep Convolutional Neural Network (2D-DCNN) and One-Dimensional Deep Convolutional Neural Network (1D-DCNN) techniques were used to extract features from panoramic radiographs and patient records. To perform age estimation using feature information, Genetic algorithm (GA) and Random Forest algorithm (RF) were modified, combined, and defined as Modified Genetic-Random Forest Algorithm (MG-RF). The performance of the system used in our study was analyzed based on the MSE, MAE, RMSE, and R2 values calculated during the implementation of the code. Results: As a result of the applied algorithms, the MSE value was 0.00027, MAE value was 0.0079, RMSE was 0.0888, and R2 score was 0.999. Conclusions: The findings of our study indicate that the AI-based system employed herein is an effective tool for age detection. Consequently, we propose that this technology could be utilized in forensic sciences in the future.
Tailored Callosotomy in Third Ventricle Colloid Cyst Resection via Anterior Interhemispheric Transcallosal Approach
(2025) Özöner, Barış; Gürses, Muhammet Enes; Öztürk, Mehmet; Arslan, Safa; Ergen, Anıl; Tubbs, Richard S.; Gonzalez-Lopez, M. Pablo; Luzzi, Sabino; Güngör, Abuzer
Background: The colloid cyst represents a relatively uncommon intracranial lesion. It garners significant attention from neurosurgeons due to its benign nature, deep-seated location, and promising prognosis when identified early and surgically removed. A variety of surgical methods are used to treat these cysts, each with their strengths and weaknesses.
Objectives: The aim of this study to introduce and assess a precise microsurgical technique for managing colloid cysts using the anterior interhemispheric transcallosal approach.
Methods: The research involved a retrospective analysis of 14 cases between 2021 and 2023 treated with the anterior interhemispheric transcallosal approach by two experienced skull base surgeons. The evaluation encompassed demographic, clinical, radiological, histological, and surgical data. Additionally, the Colloid Cyst Risk Score (CCRS) was used to assess the risk of obstructive hydrocephalus. The procedure incorporated neuronavigation and ultrasound to determine the precise entry point and to plan the trajectory.
Results: The minimally invasive microsurgical technique was effectively employed in all 14 cases, with no reported postoperative complications. Post-surgery MRI scans confirmed complete cyst removal, with an average callosotomy measurement of 5.4 ± 2.5 mm. Importantly, none of the patients experienced disconnection syndrome associated with callosotomy.
Conclusions: The adapted microsurgical approach via the anterior interhemispheric transcallosal method emerges as a secure and efficient way to address colloid cysts. It ensures comprehensive cyst removal while minimizing complications, boasting advantages such as reduced invasiveness, enhanced visibility, and minimal tissue disturbance, thereby confirming its role in colloid cyst surgical interventions.
Performance analysis of input power variations in high data rate DWDM-FSO systems under various rain conditions
(Springer, 2025) Abdulwahid, Maan Muataz; Kurnaz, Sefer; Kurnaz Türkben, Ayça; Hayal, Mohammed R.; Elsayed, Ebrahim E.; Juraev, Davron Aslonqulovich
This paper investigates the performance of a 32-channel Dense Wavelength Division Multiplexing Free-Space Optical (DWDM-FSO) system under various rain conditions and transmission distances ranging from 5 to 20 km. The study aims to identify optimal input power levels across different rain scenarios (-10 dBm, -5 dBm, 0 dBm, 5 dBm, and 10 dBm) to enhance the reliability and efficiency of optical communication in adverse weather. Findings indicate that for light rain conditions, input power levels of -10 dBm are suitable for distances up to 15 km. In moderate rain scenarios, -5 dBm is optimal for reliable communication up to 10 km, while higher input powers of 5 dBm are necessary to maintain performance in heavy rain conditions beyond 5 km. This study highlights the critical relationship between input power and atmospheric conditions, confirming that higher power levels can effectively mitigate the effects of rain-induced attenuation and scattering. Key parameters such as transmitter and receiver configurations, atmospheric attenuation, scattering, and turbulence were analyzed, demonstrating the importance of selecting appropriate power levels to ensure successful data transmission. Additionally, the research suggests future explorations into adaptive modulation techniques and quantum applications to further enhance system resilience and performance. The results provide valuable insights for system designers, enabling the adaptation of FSO systems to meet the challenges posed by varying environmental conditions and guiding developments in robust optical communication technologies.
A hybrid model using 1D-CNN with Bi-LSTM, GRU, and various ML regressors for forecasting the conception of electrical energy
(World Scientific Publishing, 2025) Abdulameer, Yahya Hafedh; Ibrahim, Abdullahi Abdu
To solve power consumption challenges by using the power of Artificial Intelligence (AI) techniques, this research presents an innovative hybrid time series forecasting approach. The suggested model combines GRU-BiLSTM with several regressors and is benchmarked against three other models to guarantee optimum reliability. It uses a specialized dataset from the Ministry of Electricity in Baghdad, Iraq. For every model architecture, three optimizers are tested: Adam, RMSprop and Nadam. Performance assessments show that the hybrid model is highly reliable, offering a practical option for model-based sequence applications that need fast computation and comprehensive context knowledge. Notably, the Adam optimizer works better than the others by promoting faster convergence and obstructing the establishment of local minima. Adam modifies the learning rate according to estimates of each parameter's first and second moments of the gradients separately. Furthermore, because of its tolerance for outliers and emphasis on fitting within a certain margin, the SVR regressor performs better than stepwise and polynomial regressors, obtaining a lower MSE of 0.008481 using the Adam optimizer. The SVR's regularization also reduces overfitting, especially when paired with Adam's flexible learning rates. The research concludes that the properties of the targeted dataset, processing demands and job complexity should all be considered when selecting a model and optimizer.
