Int J Performability Eng ›› 2022, Vol. 18 ›› Issue (10): 720-729.doi: 10.23940/ijpe.22.10.p5.720-729

Previous Articles     Next Articles

Detection, Localization and Classification of Fetal Brain Abnormalities using YOLO v4 Architecture

N. Suresh Kumar*, and Amit Kumar Goel   

  1. School of Computing Science and Engineering, Galgotias University, Uttar Pradesh, 203201, India
  • Submitted on ; Revised on ; Accepted on
  • Contact: *E-mail address: sureshkumar@galgotiasuniversity.edu.in

Abstract: In the medical field, the intensifying use of Artificial Intelligence and Machine Learning tends to lower the time and increase accuracy in detecting illnesses in their early stages. AI will have a 10.4 percent influence on the Indian economy in 2030, amounting to 0.9 trillion dollars. More than 80 anomalies in a child's fetal development have been discovered today. As it takes more time to manually identify the lesion tissues in the fetal brain, this work proposes an automatic detection and classification of fetal brain abnormalities using a cloud environment. The art of finding the abnormalities in the fetal brain is the core objective of the system, which eradicates or reduces the time and cost and improves the accuracy. The Machine Learning algorithm is introduced in fetal MRI scans to discover and pinpoint fetal brain anomalies. The Precise Epic Localization Algorithm is adopted in YOLO v4 architecture to detect and classify the healthy fetal brain with its orientation and unhealthy brain with their abnormalities from the given input of MRI Images. In this proposed work, the detection and classification of Encephalocele and Arteriovenous Malformation from a fetal brain MRI are obtained and evaluated using a machine learning algorithm to determine the abnormalities with the accuracy of 97.27%, which outperforms the public tools, BET and ROBEX. As Tesla P100 GPU is employed in the cloud environment, the output is more convenient and accessible than the existing methods.

Key words: fetal brain detection, fetal brain classification, fetal brain localization, precise epic localization (PEL) algorithm