Yazar "Alaca, Y" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • [ X ]
    Öğe
    Machine learning via DARTS-Optimized MobileViT models for pancreatic Cancer diagnosis with graph-based deep learning
    (BMC, 2025) Alaca, Y
    The diagnosis of pancreatic cancer presents a significant challenge due to the asymptomatic nature of the disease and the fact that it is frequently detected at an advanced stage. This study presents a novel approach combining graph-based data representation with DARTS-optimised MobileViT models, with the objective of enhancing diagnostic accuracy and reliability. The images of the pancreatic CT were transformed into graph structures using the Harris Corner Detection algorithm, which enables the capture of complex spatial relationships. Subsequently, the graph representations were processed using MobileViT models that had been optimised with Differentiable Architecture Search (DARTS), thereby enabling dynamic architectural adaptation. To further enhance classification accuracy, advanced machine learning algorithms, including K-Nearest Neighbours (KNN), Support Vector Machines (SVM), Random Forest (RF), and XGBoost, were applied. The MobileViTv2_150 and MobileViTv2_200 models demonstrated remarkable performance, with an accuracy of 97.33% and an F1 score of 96.25%, surpassing the capabilities of traditional CNN and Vision Transformer models. This innovative integration of graph-based deep learning and machine learning techniques demonstrates the potential of the proposed method to establish a new standard for early pancreatic cancer diagnosis. Furthermore, the study highlights the scalability of this approach for broader applications in medical imaging, which could lead to improved patient outcomes.
  • [ X ]
    Öğe
    Pancreatic Tumor Detection From CT Images Converted to Graphs Using Whale Optimization and Classification Algorithms With Transfer Learning
    (WILEY, 2025) Alaca, Y; Akmeşe, ÖF
    Pancreatic cancer is one of the most aggressive types of cancer, known for its high mortality rate, as it is often diagnosed at an advanced stage. Early diagnosis holds the potential to prolong patients' lifespans and improve treatment success rates. In this study, an innovative method is proposed to enhance the diagnosis of pancreatic cancer. Computed tomography (CT) images were converted into graphs using the Harris Corner Detection Algorithm and analyzed using deep learning models via transfer learning. DenseNet121 and InceptionV3 transfer learning models were trained on graph-based data, and model parameters were optimized using the Whale Optimization Algorithm (WOA). Additionally, classification algorithms such as k-Nearest Neighbors (k-NN), Support Vector Machines (SVM), and Random Forests (RF) were integrated into the analysis of the extracted features. The best results were achieved using the k-NN classification algorithm on features optimized by WOA, yielding an accuracy of 92.10% and an F1 score of 92.74%. The study demonstrated that graph-based transformation enabled more effective modeling of spatial relationships, thereby enhancing the performance of deep learning models. WOA offered significant superiority compared to other methods in parameter optimization. This study aims to contribute to the development of a reliable diagnostic system that can be integrated into clinical applications. In the future, the use of larger and more diverse datasets, along with different graph-based methods, could enhance the generalizability and performance of the proposed approach. The proposed model has the potential to serve as a decision support tool for physicians, particularly in early diagnosis, offering an opportunity to improve patients' quality of life.