Balanced Chest X-ray Dataset Using DCGAN and Augmentation

Balanced Dataset using DCGAN Final is a collection of chest X-ray images. The dataset was created using a Deep Convolutional Generative Adversarial Network (DCGAN) and other augmentation techniques to address class imbalance. The author, platform, and specific scale of the dataset are not provided.

Use Cases

• Train image classification models for chest pathology detection based on synthetic X-ray images.
• Benchmark generative models for medical data augmentation based on the described DCGAN methodology.
• Study the effects of synthetic data on model performance for imbalanced medical datasets.

Strengths

• Dataset is explicitly described as balanced, which can mitigate common class imbalance issues in medical imaging.
• Creation methodology is specified, involving DCGAN and augmentation techniques.

Limitations

• Row count and dataset size are unknown, which may limit suitability assessment.
• Column-level documentation and sample data are absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Kaggle

Collection Method

Created using a Deep Convolutional Generative Adversarial Network (DCGAN) and augmentation techniques.