DiGAN Breakthrough: Advancing diabetic data analysis with innovative GAN-based imbalance correction techniques
Published in Computer Methods and Programs in Biomedicine Update, 2024
This paper uniquely applies Generative Adversarial Network, traditionally used in image processing, to diabetes data analysis and classification, achieving a weighted F1 score of 90%, a 20% improvement over traditional methods.
Recommended citation: Zhao, P., Liu, X., Yue, Z., Zhao, Q., Liu, X., Deng, Y., & Wu, J. (2024). DiGAN Breakthrough: Advancing diabetic data analysis with innovative GAN-based imbalance correction techniques. Computer Methods and Programs in Biomedicine Update, 5, 100152. http://qianyuzhao.github.io/files/DiGAN.pdf