LEGAN: A Light and Effective Generative Adversarial Network for medical image synthesis

Gao, Jing; Zhao, Wenhan; Huang, Wei; Chen, Zhikui

doi:10.1016/j.compbiomed.2022.105878

Cited by 8 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach [17] can be used to optimize the projection number of CT for specific clinical applications and can significantly enhance results for cases with noise. In the paper [18], a novel method named Light and Effective Generative Adversarial Network (LEGAN) is introduced to generate high-quality medical images in a lightweight manner. The primary aim of this approach is to improve clinical diagnosis by providing additional pathological information.…”

Section: Related Literaturementioning

confidence: 99%

“…Conventional methods for medical image synthesis have certain limitations such as a lack of sensitivity towards local tissue details and the requirement of significant computing resources. To overcome these challenges, LEGAN employs a two-stage generative adversarial network with a coarse-to-fine paradigm, inspired by the painting process of humans [18]. This approach ensures the sensitivity of the model towards local information of medical images.…”

Section: Related Literaturementioning

confidence: 99%

“…This approach ensures the sensitivity of the model towards local information of medical images. Additionally, a low-rank convolutional layer is used to reduce model redundancy, which involves utilizing principal components of full-rank convolutional kernels [18]. The paper [19] illustrates the advantages and applications of GAN models in medical image fusion and discusses the challenges faced by GANs and their applications in this field.…”

Section: Related Literaturementioning

confidence: 99%

See 2 more Smart Citations

An Empirical Analysis of Diffusion, Autoencoders, and Adversarial Deep Learning Models for Predicting Dementia Using High-Fidelity MRI

Gajjar,

Garg,

Desai

et al. 2024

IEEE Access

View full text Add to dashboard Cite

This study explores cutting-edge computational technologies and intelligent methods to create realistic synthetic data, focusing on dementia-centric Magnetic Resonance Imaging (MRI) scans related to Alzheimer's and Parkinson's diseases. The research delves into Generative Adversarial Networks (GANs), Variational Autoencoders, and Diffusion Models, comparing their efficacy in generating synthetic MRI scans. Using datasets from Alzheimer's and Parkinson's patients, the study reveals intriguing findings. In the Alzheimer dataset, diffusion models produced non-dementia images with the lowest Frechet Inception Distance (FID) score at 92.46, while data-efficient GANs excelled in generating dementia images with an FID score of 178.53. In the Parkinson dataset, data-efficient GANs achieved remarkable FID scores of 102.71 for dementia images and 129.77 for non-dementia images. The study also introduces a novel aspect by incorporating a classification study, validating the generative metrics. DenseNets, a deep learning architecture, exhibited superior performance in disease detection compared to ResNets. Training both models on images generated by diffusion models further improved results, with DenseNet achieving accuracies of 80.84% and 92.42% in Alzheimer's and Parkinson's disease detection, respectively. The research not only presents innovative generative architectures but also emphasizes the importance of classification metrics, providing valuable insights into the synthesis and detection of neurodegenerative diseases through advanced computational techniques.

show abstract

Section: Related Literaturementioning

confidence: 99%

Section: Related Literaturementioning

confidence: 99%