Evaluation of Big Data based CNN Models in Classification of Skin Lesions with Melanoma

TL;DR

This paper evaluates the effectiveness of big data-trained CNN models, especially re-trained pre-existing models, for classifying skin lesions including melanoma, showing improved accuracy over models trained from scratch.

Contribution

It demonstrates that re-training pre-trained CNN models on dermatoscopic images enhances skin lesion classification accuracy, with ResNet-50 achieving 93.89% accuracy.

Findings

Re-trained CNN models outperform models trained from scratch.

ResNet-50 achieved 93.89% accuracy in skin lesion classification.

Classification accuracy for melanoma and basal cell carcinoma was 79.13% and 82.88%, respectively.

Abstract

This chapter presents a methodology for diagnosis of pigmented skin lesions using convolutional neural networks. The architecture is based on convolu-tional neural networks and it is evaluated using new CNN models as well as re-trained modification of pre-existing CNN models were used. The experi-mental results showed that CNN models pre-trained on big datasets for gen-eral purpose image classification when re-trained in order to identify skin le-sion types offer more accurate results when compared to convolutional neural network models trained explicitly from the dermatoscopic images. The best performance was achieved by re-training a modified version of ResNet-50 convolutional neural network with accuracy equal to 93.89%. Analysis on skin lesion pathology type was also performed with classification accuracy for melanoma and basal cell carcinoma being equal to 79.13% and 82.88%, respectively.