Segmentation Of Intracerebral Haemorrhage CT-Scan Images Using N-Net Architecture

In recent years, image segmentation has emerged as a critical task in computer vision and various applications such as medical image processing, autonomous driving, and satellite imagery analysis. Deep learning techniques, particularly convolutional neural networks (CNNs), have played a pivotal role in revolutionizing image segmentation by offering unprecedented accuracy and efficiency. Many algorithms have used for the segmentation such as U-Net, V-Mesh, U-Net++ and etc. But, U-Net having the challenges in parameter transfer, model size and in speed. To overcome these issues, this paper explores the evolution of image segmentation techniques into ICH (Intracerebral Haemorrhage) for better diagnosis. This model is tested using ICH dataset available in Kaggle. In which, N-Net have produced the better performance in terms of precision, Dice, Recall and F1 Score.