Yazar "Savas, Serkan" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • [ X ]
    Öğe
    Enhancing Disease Classification with Deep Learning: a Two-Stage Optimization Approach for Monkeypox and Similar Skin Lesion Diseases
    (Springer, 2024) Savas, Serkan
    Monkeypox (MPox) is an infectious disease caused by the monkeypox virus, presenting challenges in accurate identification due to its resemblance to other diseases. This study introduces a deep learning-based method to distinguish visually similar diseases, specifically MPox, chickenpox, and measles, addressing the 2022 global MPox outbreak. A two-stage optimization approach was presented in the study. By analyzing pre-trained deep neural networks including 71 models, this study optimizes accuracy through transfer learning, fine-tuning, and ensemble learning techniques. ConvNeXtBase, Large, and XLarge models were identified achieving 97.5% accuracy in the first stage. Afterwards, some selection criteria were followed for the models identified in the first stage for use in ensemble learning technique within the optimization approach. The top-performing ensemble model, EM3 (composed of RegNetX160, ResNetRS101, and ResNet101), attains an AUC of 0.9971 in the second stage. Evaluation on unseen data ensures model robustness and enhances the study's overall validity and reliability. The design and implementation of the study have been optimized to address the limitations identified in the literature. This approach offers a rapid and highly accurate decision support system for timely MPox diagnosis, reducing human error, manual processes, and enhancing clinic efficiency. It aids in early MPox detection, addresses diverse disease challenges, and informs imaging device software development. The study's broad implications support global health efforts and showcase artificial intelligence potential in medical informatics for disease identification and diagnosis.
  • [ X ]
    Öğe
    Ensemble learning based lung and colon cancer classification with pre-trained deep neural networks
    (Springer Heidelberg, 2024) Savas, Serkan; Guler, Osman
    PurposeLung and colon cancer are among the most common cancer types with the highest mortality rates worldwide. In recent years, studies have been carried out on the early diagnosis and detection of lung cancer and colon cancer. Also, new treatment methods and options have been investigated. With the development of early diagnostic methods, the survival rate of patients has increased. Therefore, in this study, a transfer learning-based deep ensemble learning model is proposed for the classification of lung and colon cancer from histopathological images.MethodsExperiments were carried out on the LC25000 dataset with the proposed approach. DenseNet121, InceptionV3, MobileNet, ResNet50, ResNet101, VGG16, and Xception models were used for transfer learning. These models were implemented with different variations for ensemble learning.ResultsIt is seen that the most successful transfer learning model is the InceptionV3 model, and the ResNet101 model has the lowest performance. The DETL_V1 model, in which all models were used, was the most successful model with an accuracy rate of 99.78%.ConclusionBy focusing on top models and using ensemble learning to combine their predictions, better performance was obtained compared to any single model on its own. Obtaining a more robust and generalizable model was achieved by combining multiple models that are trained on different subsets of the data with different architectures. The results proved that the proposed approach is supportive for decision in clinical diagnosis processes.