No abstract

Melanoma is one of the main causes of cancer-related deaths. The development of new computational methods as an important tool for assisting doctors can lead to early diagnosis and effectively reduce mortality. In this work, we propose a convolutional neural network architecture for melanoma diagnosis inspired by ensemble learning and genetic algorithms. The architecture is designed by a genetic algorithm that finds optimal members of the ensemble. Additionally, the abstract features of all models are merged and, as a result, additional prediction capabilities are obtained. The diagnosis is achieved by combining all individual predictions. In this manner, the training process is implicitly regularized, showing better convergence, mitigating the overfitting of the model, and improving the generalization performance. The aim is to find the models that best contribute to the ensemble. The proposed approach also leverages data augmentation, transfer learning, and a segmentation algorithm. The segmentation can be performed without training and with a central processing unit, thus avoiding a significant amount of computational power, while maintaining its competitive performance. To evaluate the proposal, an extensive experimental study was conducted on sixteen skin image datasets, where state-of-the-art models were significantly outperformed. This study corroborated that genetic algorithms can be employed to effectively find suitable architectures for the diagnosis of melanoma, achieving in overall 11% and 13% better prediction performances compared to the closest model in dermoscopic and non-dermoscopic images, respectively. Finally, the proposal was implemented in a web application in order to assist dermatologists and it can be consulted at http://skinensemble.com.

Skin cancer is one of the most common types of cancers in the world, with melanoma being the most lethal form. Automatic melanoma diagnosis from skin images has recently gained attention within the machine learning community, due to the complexity involved. In the past few years, convolutional neural network models have been commonly used to approach this issue. This type of model, however, presents disadvantages that sometimes hamper its application in real-world situations, e.g., the construction of transformation-invariant models and their inability to consider spatial hierarchies between entities within an image. Recently, Dynamic Routing between Capsules architecture (CapsNet) has been proposed to overcome such limitations. This work is aimed at proposing a new architecture which combines convolutional blocks with a customized CapsNet architecture, allowing for the extraction of richer abstract features. This architecture uses high-quality 299×299×3 skin lesion images, and a hyper-tuning of the main parameters is performed in order to ensure effective learning under limited training data. An extensive experimental study on eleven image datasets was conducted where the proposal significantly outperformed several state-of-the-art models. Finally, predictions made by the model were validated through the application of two modern model-agnostic interpretation tools.