Cilt:64 Sayı:01 (2022)

Permanent URI for this collection

https://dspace.ankara.edu.tr/handle/20.500.12575/86465

Browse

Browsing Cilt:64 Sayı:01 (2022) by Subject "deep learning"

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    3D reconstruction of coronary arteries using deep networks from synthetic X-ray angiogram data
    (Ankara Üniversitesi Mühendislik Fakültesi, 2022) Atlı, İbrahim; Other; Other
    Cardiovascular disease (CVD) is one of the most common health problems that are responsible for one-third of all deaths around the globe. Although X-Ray angiography has deficiencies such as two-dimensional (2D) representation of three dimensional (3D) structures, vessel overlapping, noisy background, the existence of other tissues/organs in images, etc., it is used as the gold standard technique for the diagnosis and in some cases treatment of CVDs. To overcome the deficiencies, great efforts have been drawn on retrieval of actual 3D representation of coronary arterial tree from 2D X-ray angiograms. However, the proposed algorithms are based on analytical methods and enforce some constraints. With the evolution of deep neural networks, 3D reconstruction from images can be achieved effectively. In this study, we propose a new data structure for the representation of objects in a tubular shape for 3D reconstruction of arteries using deep learning. Moreover, we propose a method to generate synthetic coronaries from data of real subjects. Then, we validate tubular shape representation using 3 typical deep learning architectures with synthetic X-ray data we produced. The input to deep learning architectures is multi-view segmented X-Ray images and the output is the structured tubular representation. We compare results qualitatively in terms of visual appearance and quantitatively in terms of Chamfer Distance and Mean Squared Error. The results demonstrate that tubular representation has promising performance in 3D reconstruction of coronaries. We observe that convolutional neural network (CNN) based architectures yield better 3D reconstruction performance with 9.9e-3 on Chamfer Distance. On the other hand, LSTM-based network fails to learn the coronary tree structure and we conclude that LSTMs are not appropriate for auto-regression problems as depicted in this study.
  • No Thumbnail Available
    Item
    Utilization of deep learning architectures for MIMO detection
    (Ankara Üniversitesi Mühendislik Fakültesi, 2022) Karahan, Sümeyye Nur; Elektrik-Elektronik Mühendisliği; Mühendislik Fakültesi
    Applications of deep learning in communications systems are becoming popular today with their powerful solutions to complex problems. This study considers the utilization of deep learning detectors for small-scale multiple-input multiple-output systems. Deep neural network, long short-term memory, and one-dimenisonal convolutional neural network architectures are discussed and the bit error rate performances of these deep learning based detectors are compared with the optimal maximum likelihood and sub-optimal minimum mean square error detectors. Simulation results show that the deep neural network architecture has the best detection performance among the discussed deep learning detectors and may outperform the sub-optimal minimum mean square error detector. For small-scale multiple-input multiple-output systems, the performance of the deep learning based detector is close to that of the optimal detector.