Evaluating the Clinical Realism of Synthetic Chest X-Rays Generated Using Progressively Growing GANs

Chest x-rays are a vital tool in the workup of many patients. Similar to most medical imaging modalities, they are profoundly multi-modal and are capable of visualising a variety of combinations of conditions... (read more)

PDF Abstract

Datasets


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Medical Image Generation ChestXray14 1024x1024 Progressive Growing GAN FID 8.02 # 1

Methods used in the Paper


METHOD TYPE
Batch Normalization
Normalization
Concatenated Skip Connection
Skip Connections
ReLU
Activation Functions
Latent Optimisation
Latent Variable Sampling
Dropout
Regularization
Kaiming Initialization
Initialization
Global Average Pooling
Pooling Operations
Average Pooling
Pooling Operations
Max Pooling
Pooling Operations
Dense Block
Image Model Blocks
DenseNet
Convolutional Neural Networks
WGAN-GP Loss
Loss Functions
1x1 Convolution
Convolutions
Dense Connections
Feedforward Networks
Leaky ReLU
Activation Functions
Convolution
Convolutions
Local Response Normalization
Normalization
ProGAN
Generative Models