Browsing by Author "Ersoy E.O."

Now showing 1 - 2 of 2

Accurate Dictionary Matching for Mr Fingerprinting Using Neural Networks and Feature Extraction
(Institute of Electrical and Electronics Engineers Inc., 2020) Soyak R.; Ersoy E.O.; Navruz E.; Fakultesi M.; Unay D.; Oksuz I.
Magnetic Resonance Fingerprinting is a recent technique which aims at providing simultaneous measurements of multiple parameters. MRF works by varying acquisition parameters in a pseudorandom manner so as to get unique, uncorrelated signal evolutions from each tissue. MRF is a dictionary based approach, and thus requires a database. This database can be created by simulating the signal evolutions from first principles using different physical models for a wide variety of tissue parameter combinations. Having this dictionary, a pattern recognition algorithm is used to match the acquired signal evolutions from each voxel with each signal evolution in the dictionary. In this paper, we compare the efficiency of deep learning based feature extraction method and neural network architectures in order to achieve state-of-the-art accuracy in dictionary matching for MRF. Our results showcase successful dictionary matching with high accuracy both quantitatively and qualitatively. © 2020 IEEE.
Citation - Scopus: 1
Binocular Vision Based Convolutional Networks
(Institute of Electrical and Electronics Engineers Inc., 2020) Oktar Y.; Ulucan O.; Karakaya D.; Ersoy E.O.; Türkan, Mehmet
It is arguable that whether the single camera captured (monocular) image datasets are sufficient enough to train and test convolutional neural networks (CNNs) for imitating the biological neural network structures of the human brain. As human visual system works in binocular, the collaboration of the eyes with the two brain lobes needs more investigation for improvements in such CNN-based visual imagery analysis applications. It is indeed questionable that if respective visual fields of each eye and the associated brain lobes are responsible for different learning abilities of the same scene. There are such open questions in this field of research which need rigorous investigation in order to further understand the nature of the human visual system, hence improve the currently available deep learning applications. This paper analyses a binocular CNNs architecture that is more analogous to the biological structure of the human visual system than the conventional deep learning techniques. While taking a structure called optic chiasma into account, this architecture consists of basically two parallel CNN structures associated with each visual field and the brain lobe, fully connected later possibly as in the primary visual cortex. Experimental results demonstrate that binocular learning of two different visual fields leads to better classification rates on average, when compared to classical CNN architectures. © 2020 IEEE.