{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/skin-lesion-diagnosis-using-ensembles","title":"Skin Lesion Diagnosis using Ensembles, Unscaled Multi-Crop Evaluation and Loss Weighting","arxiv_id":"1808.01694","date":"2018-08-05","proceeding":null,"authors":["Nils Gessert","Thilo Sentker","Frederic Madesta","Rüdiger Schmitz","Helge Kniep","Ivo Baltruschat","René Werner","Alexander Schlaefer"],"abstract":"In this paper we present the methods of our submission to the ISIC 2018\nchallenge for skin lesion diagnosis (Task 3). The dataset consists of 10000\nimages with seven image-level classes to be distinguished by an automated\nalgorithm. We employ an ensemble of convolutional neural networks for this\ntask. In particular, we fine-tune pretrained state-of-the-art deep learning\nmodels such as Densenet, SENet and ResNeXt. We identify heavy class imbalance\nas a key problem for this challenge and consider multiple balancing approaches\nsuch as loss weighting and balanced batch sampling. Another important feature\nof our pipeline is the use of a vast amount of unscaled crops for evaluation.\nLast, we consider meta learning approaches for the final predictions. Our team\nplaced second at the challenge while being the best approach using only\npublicly available data.","url_abs":"http://arxiv.org/abs/1808.01694v1","url_pdf":"http://arxiv.org/pdf/1808.01694v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"skin-lesion-diagnosis-using-ensembles","repo_url":"https://github.com/ngessert/isic2018","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"skin-lesion-diagnosis-using-ensembles","repo_url":"https://github.com/ngessert/patch-lesion","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"meta-learning","task_name":"Meta-Learning"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-block","method_name":"Dense Block"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"grouped-convolution","method_name":"Grouped Convolution"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"resnext","method_name":"ResNeXt"},{"method_slug":"resnext-block","method_name":"ResNeXt Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"senet","method_name":"SENet"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"squeeze-and-excitation-block","method_name":"Squeeze-and-Excitation Block"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}