Exploring Content Based Image Retrieval for Highly Imbalanced Melanoma Data using Style Transfer, Semantic Image Segmentation and Ensemble Learning

TL;DR

This paper investigates content-based image retrieval for melanoma lesion images in open-set scenarios, proposing novel similarity measures including style loss and ensemble learning to improve accuracy over traditional methods.

Contribution

It introduces a new similarity measure called I1-Score combining style loss and Dice coefficient, and demonstrates the effectiveness of style loss in CBIR for imbalanced melanoma data.

Findings

Style loss-based similarity achieves higher accuracy than Euclidean and Cosine measures.

Ensemble learning improves model generalization.

Dice coefficient similarity performs poorly compared to style loss.

Abstract

Lesion images are frequently taken in open-set settings. Because of this, the image data generated is extremely varied in nature.It is difficult for a convolutional neural network to find proper features and generalise well, as a result content based image retrieval (CBIR) system for lesion images are difficult to build. This paper explores this domain and proposes multiple similarity measures which uses Style Loss and Dice Coefficient via a novel similarity measure called I1-Score. Out of the CBIR similarity measures proposed, pure style loss approach achieves a remarkable accuracy increase over traditional approaches like Euclidean Distance and Cosine Similarity. The I1-Scores using style loss performed better than traditional approaches by a small margin, whereas, I1-Scores with dice-coefficient faired very poorly. The model used is trained using ensemble learning for better generalization.