Generative Adversarial Networks (GAN) have shown potential in expanding limited medical imaging datasets.This study explores how different ratios of GAN-generated and real brain tumor MRI images impact the performance of a CNN in classifying healthy vs. tumorous scans.A DCGAN was used to create synthetic images which were mixed with real ones at various ratios to train a custom CNN.The CNN was then evaluated on a separate real-world test set.Our results indicate that the model maintains high sensitivity and precision in tumor classification, even when trained predominantly on synthetic data.When only a small portion of GAN data was added, such as 900 real images and 100 GAN images, the model achieved excellent performance, with test accuracy reaching 95.2%, and precision, recall, and F1-score all exceeding 95%.However, as the proportion of GAN images increased further, performance gradually declined.This study suggests that while GANs are useful for augmenting limited datasets especially when real data is scarce, too much synthetic data can introduce artifacts that affect the model's ability to generalize to real world cases.
This study investigates the application of Generative Adversarial Networks (GANs) in augmenting datasets for brain tumor classification using Convolutional Neural Networks (CNNs). The authors explored the impact of various ratios of GAN-generated and real MRI images on the classification performance of a custom CNN designed for this task. The research highlighted that initial inclusion of synthetic data improved performance, achieving an accuracy of 95.2% with a mix of 90% real and 10% GAN data. However, as the proportion of GAN images increased, the performance metrics including accuracy, precision, and recall decreased, demonstrating the issues of over-reliance on synthetic data. The study concludes that while GANs can effectively augment limited datasets, maintaining a balance between real and synthetic images is crucial for optimal model performance.
This paper employs the following methods:
- Generative Adversarial Networks
- Convolutional Neural Networks
The following datasets were used in this research:
- Accuracy
- Precision
- Recall
- F1-Score
- AUC
- Achieved test accuracy of 95.2% with 90% real and 10% GAN data
- F1-score exceeding 95% with the ideal data mix
- Performance declined with increased GAN data proportion
- Number of GPUs: None specified
- GPU Type: None specified
- Compute Requirements: trained using binary cross-entropy loss for 1000 epochs with a batch size of 64 and learning rate of 1 × 10 −4