A Scrupulous Approach to Perform Classification and Detection of Fetal Brain using Darknet YOLO v4
N. Suresh Kumar, Amit Kumar Goel, S. Jayanthi
Abstract
A term concerning the development of a non-human intelligence project is artificial intelligence (Al). According to the growth rate, by 2030, artificial intelligence would contribute more than fifteen percentage points to seven trillion to the global economy, with the most substantial effects on the healthcare industry. A convolutional neural network is a form of neural network that is most frequently applied to image processing problems. A computer recognizes artifacts in a picture and utilizes convolutional neural networks that are so essential in deep learning and artificial intelligence today. So powerful is the convolutional neural network for identifying and classifying pictures. Classification and object identification is the main objective of this research work. The darknet yolov4 is used, to perform the classification, and region of interest detection with the best accuracy scores. The model is trained with the Tesla GPU and obtained the results of the existing techniques in the field of fetal brain classification and localization. The accuracy of 97.92% and precision percentage of 96.70 is achieved in the research work.