Transfer Learning with Ensembles of Deep Neural Networks for Skin Cancer Detection in Imbalanced Data Sets
This work addresses the challenge of imbalanced data in medical image analysis for skin cancer detection, offering a potential improvement in diagnostic accuracy, though it appears incremental as it builds on existing transfer learning and ensemble methods.
The paper tackled the problem of skin cancer detection from medical images with limited and imbalanced data by proposing a novel ensemble-based CNN architecture that combines pre-trained and newly trained models with metadata, achieving favorable performance compared to seven benchmark methods in terms of F1-measure, AUC-ROC, and AUC-PR on a dataset of 33,126 images.
Several machine learning techniques for accurate detection of skin cancer from medical images have been reported. Many of these techniques are based on pre-trained convolutional neural networks (CNNs), which enable training the models based on limited amounts of training data. However, the classification accuracy of these models still tends to be severely limited by the scarcity of representative images from malignant tumours. We propose a novel ensemble-based CNN architecture where multiple CNN models, some of which are pre-trained and some are trained only on the data at hand, along with auxiliary data in the form of metadata associated with the input images, are combined using a meta-learner. The proposed approach improves the model's ability to handle limited and imbalanced data. We demonstrate the benefits of the proposed technique using a dataset with 33126 dermoscopic images from 2056 patients. We evaluate the performance of the proposed technique in terms of the F1-measure, area under the ROC curve (AUC-ROC), and area under the PR-curve (AUC-PR), and compare it with that of seven different benchmark methods, including two recent CNN-based techniques. The proposed technique compares favourably in terms of all the evaluation metrics.