Browsing by Author "Ornek, A.H."

Now showing 1 - 2 of 2

Citation - Scopus: 11
Classification of Unhealthy and Healthy Neonates in Neonatal Intensive Care Units Using Medical Thermography Processing and Artificial Neural Network
(Elsevier, 2019) Savaşcı, D.; Ornek, A.H.; Ervural, S.; Ceylan, M.; Konak, M.; Soylu, H.
Tracking temperature changes of neonatals in the neonatal intensive care unit is quite important in the prediagnosis of diseases or the evaluation of follow-up treatment. The purpose of this study is to develop an analysis system based on thermal imaging, which is the contact-free, nonionized and noninvasive method for the neonatal. For this purpose, 190 images taken from 19 healthy and 19 unhealthy neonates were used. In general, this study consists of three steps. First, the temperature map of the images was segmented. Then, discrete wavelet transform (DWT), curvelet transform and contourlet transform as multiresolution methods were applied to them, and feature vectors were extracted by using their approximation coefficients. After that, all feature vectors were given as an input to the artificial neural networks (ANN) and support vector machines. According to the obtained results, the best accuracy rate was 98.42% when using DWT+ANN. © 2019 Elsevier Inc. All rights reserved.
Citation - Scopus: 1
Haycamj: a New Method To Uncover the Importance of Main Filter for Small Objects in Explainable Artificial Intelligence
(Springer Science and Business Media Deutschland GmbH, 2024) Ornek, A.H.; Ceylan, M.
Visual XAI methods enable experts to reveal importance maps highlighting intended classes over input images. This research paper presents a novel approach to visual explainable artificial intelligence (XAI) for object detection in deep learning models. The study investigates the effectiveness of activation maps generated by five different methods, namely GradCAM, GradCAM++, EigenCAM, HayCAM, and a newly proposed method called "HayCAMJ", in detecting objects within images. The experiments were conducted on two datasets (Pascal VOC 2007 and Pascal VOC 2012) and three models (ResNet18, ResNet34, and MobileNet). Zero padding was applied to resize and center the objects due to the large objects in the images. The results show that HayCAMJ performs better than other XAI techniques in detecting small objects. This finding suggests that HayCAMJ has the potential to become a promising new approach for object detection in deep classification models. © The Author(s) 2024.