Yazar "Alajaj, Mustafa Isam" seçeneğine göre listele

Listeleniyor 1 - 1 / 1
  • Küçük Resim
    Yayın
    Automated detection and classification of spinal disc herniation using deep learning on MRI images
    (İstanbul Gedik Üniversitesi, Lisansüstü Eğitim Enstitüsü, 2024) Alajaj, Mustafa Isam; Dilek, Rıza
    A common pathological condition that affects the intervertebral discs is spinal disc herniation. Nerve compression and pain result when the gel-like substance in a disc protrudes through the stiff outer layer. For ideal treatment arranging and patient consideration, an exact and provoke finding of spinal disc herniation should be made. Spinal disc herniation necessitates prompt and precise diagnosis before the appropriate treatment can be planned. Magnetic Resonance Imaging (MRI) has developed as an efficient imaging technique in the diagnosis and evaluation of spinal disc herniation, offering precise anatomical information and allowing for non-invasive examination. While MRI is an efficient imaging method for identifying spinal disc herniation, radiologists' manual review procedure is time- consuming and error-prone. Development of an automated solution include the spine's complex anatomical structure, imaging technique variability, and the lack of annotated datasets raise significant challenges. In the last few years, deep learning algorithms have transformed the interpretation of medical imaging, demonstrating remarkable ability in automated tasks like as detection and classification. So, a computer system that uses deep learning techniques to identify and classify spinal disc herniation on MRI images is required. These strategies are superior to traditional computer-aided diagnostic approaches because they can instantly acquire specific patterns and features from large data sets. However, there are many difficulties in creating a deep learning system that is both reliable and effective for distinguishing intervertebral disc herniation. The complex anatomy of the spine, the unpredictable nature of imaging strategies, and powerful algorithms that can handle the occurrence of abnormalities on MRI images. Also, one more obstruction to effectively training deep learning models is the absence of large, annotated datasets. Hence, it is necessary to discover how to develop a deep learning system that overcomes those challenges and uses MRI scans to accurately and automatically identify and classify spinal disc herniations. This research developed a deep learning system specifically designed for the automated detection and categorization of spinal disc herniation using MRI data, for identifying and categorizing spinal disc herniation from MRI data by utilizing deep learning capabilities. The research compared the efficiency of the conventional approaches to computer-aided diagnosis to the deep learning-based method. In addition to surveying the robotized framework's clinical helpfulness by assessing its ability to help radiologists in deciding the right course of treatment and the board of spinal disc herniation. By applying such deep learning system into a therapeutically useful tool, radiologists can use it to correctly diagnose and classify spinal disc herniation from AUTOMATED DETECTION AND CLASSIFICATION OF SPINAL DISC HERNIATION USING DEEP LEARNING ON MRI IMAGES MRI data. The viable reception of such a system can possibly work on indicative interaction, treatment arranging, and patient results. By demonstrating the efficacy of deep learning techniques in improving the diagnosis and treatment of spinal herniation of the disc, the research contributed to the advancement of computer-aided medical image analysis. This study can possibly change the area of spinal disc herniation discovery and add to the development of PC supported examination of clinical pictures by utilizing the force of deep learning.