In Journal of healthcare informatics research
A novel approach of data augmentation based on irregular superpixel decomposition is proposed. This approach called SuperpixelGridMasks permits to extend original image datasets that are required by training stages of machine learning-related analysis architectures towards increasing their performances. Three variants named SuperpixelGridCut, SuperpixelGridMean, and SuperpixelGridMix are presented. These grid-based methods produce a new style of image transformations using the dropping and fusing of information. Extensive experiments using various image classification models as well as precision health and surrounding real-world datasets show that baseline performances can be significantly outperformed using our methods. The comparative study also shows that our methods can overpass the performances of other data augmentations. SuperpixelGridCut, SuperpixelGridMean, and SuperpixelGridMix codes are publicly available at https://github.com/hammoudiproject/SuperpixelGridMasks.
Hammoudi Karim, Cabani Adnane, Slika Bouthaina, Benhabiles Halim, Dornaika Fadi, Melkemi Mahmoud
2023-Jan-13
Data analytics, Data augmentation, Deep learning, Health informatics, Medical scans, Precision health, Predictive classification model, Real-world applications, Wellness
