Abstract: Fetal brain abnormalities are a major concern in prenatal healthcare, often resulting in significant neurodevelopmental complications. This study presents an innovative approach to detecting 14 specific fetal brain abnormalities using the YOLOv5 (You Only Look Once version 5) deep learning model. By applying this state-of-the- art object detection algorithm to medical imaging, including ultrasound and MRI scans, the project aims to accurately identify abnormalities such as hydrocephalus, neural tube defects, and various cerebral malformations. YOLOv5’s ability to provide real-time, high-accuracy detection allows for early diagnosis and intervention, significantly enhancing prenatal care. The research details the dataset used, the training process, and the evaluation of YOLOv5’s performance in terms of precision, recall, and overall detection accuracy. The results highlight the potential of deep learning, specifically YOLOv5, in revolutionizing the diagnostic process for fetal brain abnormalities, leading to improved clinical outcomes and better management strategies for at-risk pregnancies.
|
DOI:
10.17148/IJARCCE.2024.131234
