Özyazici, Kaan

Profile Picture
Profile URL
https://gcris.yasar.edu.tr/handle/123456789/13067
Name Variants
Job Title
Araş.Gör.
Email Address
Main Affiliation
01.01.12.03. Yönetim Bilişim Sistemleri Bölümü
Status
Current Staff
Website
ORCID ID
0000-0002-4836-8352
Scopus Author ID
57214806539
Turkish CoHE Profile ID
Google Scholar ID
WoS Researcher ID
FQU-4603-2022
Files
kaan.ozyazici.jpg (7.17 KB)

Sustainable Development Goals

SDG data is not available
Documents

2

Citations

78

h-index

1

Go to Scopus profile
Documents

2

Citations

54

Go to WoS profile
Scholarly Output

2

Articles

2

Views / Downloads

0/0

Supervised MSc Theses

0

Supervised PhD Theses

0

WoS Citation Count

55

Scopus Citation Count

79

Patents

0

Projects

0

WoS Citations per Publication

27.50

Scopus Citations per Publication

39.50

Open Access Source

1

Supervised Theses

0

JournalCount
Engineering Science and Technology, an International Journal1
Microscopy Research and Technique1
Current Page: 1 / 1

Scopus Quartile Distribution

Quartile distribution chart data is not available

Competency Cloud

GCRIS Competency Cloud

Filters

2020 - 20262
Reset filters

Settings

Scholarly Output Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Article
    Citation - WoS: 55
    Citation - Scopus: 79
    Implementation of machine learning algorithms for gait recognition
    (ELSEVIER - DIVISION REED ELSEVIER INDIA PVT LTD, 2020) Aybuke Kececi; Armagan Yildirak; Kaan Ozyazici; Gulsen Ayluctarhan; Onur Agbulut; Ibrahim Zincir; Ozyazici, Kaan; Kececi, Aybuke; Ayluctarhan, Gulsen; Zincir, Ibrahim; Yildirak, Armagan; Agbulut, Onur
    The basis of biometric authentication is that each person's physical and behavioural characteristics can be accurately defined. Many authentication techniques were developed over the years. Human gait recognition is one of these techniques. This article explores machine learning techniques for user authentication on HugaDB database which is a human gait data collection for analysis and activity recognition (Chereshnev and Kertesz-Farkas 2017). The activities recorded in this dataset are walking running sitting and standing. The data were collected with devices such as wearable accelerometer and gyroscope. In total the data describe 18 individuals thus we considered each individual as a different class. 10 commonly used machine learning algorithms have been implemented over the HugaDB. The proposed system achieved more than 99% in accuracy via IB1 Random Forest and Bayesian Net algorithms. (C) 2020 Karabuk University. Publishing services by Elsevier B.V.
  • Thumbnail Image
    Article
    HistoNeRF: An Accessible and Intelligent Approach for Comprehensive 2D-to-3D Histological Assessment
    (Wiley, 2026) Kilic, Kubilay Dogan; Ozyazici, Kaan; Yilmaz, Zeynep Simge; Ozyazici, Aysegul Taskiran; Horuz, Busra; Taşkıran Özyazici, Ayşegül; Kisaoglu, Huseyin
    Histological analysis is central to biomedical research and diagnostic pathology, yet conventional two-dimensional (2D) sectioning captures only limited aspects of tissue architecture. Critical spatial relationships-such as tumor boundaries, stromal organization, and vascular networks-remain obscured, restricting diagnostic accuracy and biological interpretation. HistoNeRF addresses these limitations by adapting Neural Radiance Fields (NeRF) to reconstruct three-dimensional (3D) tissue volumes from routine histological sections. In this study, 84 toluidine blue (TB)-stained murine ovarian sections were digitized, alignment-corrected, and integrated into volumetric models. Tissue segmentation was performed using a convolutional neural network, while visualization was achieved through an interactive, GPU-accelerated interface. To ensure accessibility and reproducibility, a Python-based graphical application (HistoNeRF GUI) was developed following Human-Computer Interaction (HCI) principles and containerized with Docker, allowing installation-free deployment via Docker Hub. HistoNeRF produced high-fidelity 3D reconstructions (SSIM = 0.92; Dice similarity coefficient = 0.88), enabling expert histologists to better visualize follicular structures, stromal compartments, and vascular elements. The containerized GUI was deployed successfully from Docker Hub, providing immediate access to 3D reconstruction without a complex local setup. By overcoming the inherent constraints of 2D microscopy, HistoNeRF enhances the visualization, interpretability, and reproducibility of histological architecture. The HCI-guided, cross-platform interface supports scalability and rapid adoption in digital pathology workflows. Although validation was limited to murine ovarian tissue and one staining protocol, this framework can be extended across tissue types and clinical datasets. HistoNeRF bridges routine histology and 3D volumetric analysis through accurate, interactive reconstructions that advance diagnostic precision and biomedical research. While demonstrated on 84 serial TB-stained ovarian sections, broader validation across tissues, stains, and pathological conditions remains future work; to support this, we provide a Dockerized, modular pipeline for straightforward extension.