A Novel Generative Adversarial Network-Based Approach for Automated Brain Tumour Segmentation

Abstract

Background: Medical image segmentation is more complicated and demanding than ordinary image segmentation due to the density of medical pictures. A brain tumour is the most common cause of high mortality. Objectives: Extraction of tumorous cells is particularly difficult due to the differences between tumorous and non-tumorous cells. In ordinary convolutional neural networks, local background information is restricted. As a result, previous deep learning algorithms in medical imaging have struggled to detect anomalies in diverse cells. Methods: As a solution to this challenge, a deep convolutional generative adversarial network for tumour segmentation from brain Magnetic resonance Imaging (MRI) images is proposed. A generator and a discriminator are the two networks that make up the proposed model. This network focuses on tumour localisation, noise-related issues, and social class disparities. Results: Dice Score Coefficient (DSC), Peak Signal to Noise Ratio(PSNR), and Structural Index Similarity (SSIM) are all generally 0.894, 62.084 dB, and 0.88912, respectively. The model’s accuracy has improved to 97 percent, and its loss has reduced to 0.012. Conclusions: Experiments reveal that the proposed approach may successfully segment tumorous and benign tissues. As a result, a novel brain tumour segmentation approach has been created.

Anomaly Detection in Medical Time Series with Generative Adversarial Networks: A Selective Review

“…The process of applying GANs to image segmentation is slightly different than that of applying vanilla GAN: the generator now aims to create an image where each pixel corresponds to a particular class label and the discriminator attempts to differentiate between the ground-truth segmentation (real) and the generator's segmentation (synthetic). This is again a type of image-to-image translation for which GANs have been used successfully [100][101][102].…”

Improving the Generalizability of Deep Learning for T2-Lesion Segmentation of Gliomas in the Post-Treatment Setting

“…Currently, the most prevalent and high-performing methods in the computer vision literature for automated tumor segmentation use a variety of encoder-decoder-based architectures in an end-to-end approach to generate lesion masks directly from MR image inputs [30,32,35]. These methods are primarily inspired by the pioneering developments of convolutional neural networks (CNNs) capable of 3D segmentation in works like U-net [19] and V-net [20], with more recent advances employing techniques such as multi-task learning [22][23][24], generative modeling for augmenting training data or adversarial approaches [26,[36][37][38][39][40][41], hybrid machine learning approaches [27,42], domain adaptation and transfer learning [29,[43][44][45][46][47][48][49][50][51][52][53], task-specific loss modification [18,25,27,31,34,54], diffusion models [41,[55][56][57], and attention mechanisms like transformer modules [58][59][60], as well as federated learning approaches [34,...…”