Bulk Production Augmentation Towards Explainable Melanoma Diagnosis
This addresses the need for diagnostic evidence in melanoma diagnosis for medical applications, but it is incremental as it builds on existing data augmentation methods.
The paper tackles the problem of limited reliable training data for explainable melanoma diagnosis by proposing bulk production augmentation (BPA) to generate high-quality pseudo-skin tumor images, which boosts atypical pigment network (APN) detection performance by 20.0 percentage points in AUC.
Although highly accurate automated diagnostic techniques for melanoma have been reported, the realization of a system capable of providing diagnostic evidence based on medical indices remains an open issue because of difficulties in obtaining reliable training data. In this paper, we propose bulk production augmentation (BPA) to generate high-quality, diverse pseudo-skin tumor images with the desired structural malignant features for additional training images from a limited number of labeled images. The proposed BPA acts as an effective data augmentation in constructing the feature detector for the atypical pigment network (APN), which is a key structure in melanoma diagnosis. Experiments show that training with images generated by our BPA largely boosts the APN detection performance by 20.0 percentage points in the area under the receiver operating characteristic curve, which is 11.5 to 13.7 points higher than that of conventional CycleGAN-based augmentations in AUC.