Deep Learning, Sparse Coding, and SVM for Melanoma Recognition in Dermoscopy Images

By Noel C. F. Codella, Junjie Cai, Mani Abedini, Rahil Garnavi, A. Halpern, John R. Smith. Machine Learning for Multimodal Interaction, 2015

379 citations recorded by the source corpus.

This work presents an approach for melanoma recognition in dermoscopy images that combines deep learning, sparse coding, and support vector machine SVM learning algorithms. One of the beneficial aspects of the proposed approach is that unsupervised learning within the domain, and feature transfer from the domain of natural photographs, eliminates the need of annotated data in the target task to learn good features. The applied feature transfer also allows the system to draw analogies between observations in dermoscopic images and observations in the natural world, mimicking the process clinical experts themselves employ to describe patterns in skin lesions. To evaluate the methodology, performance is measured on a dataset obtained from the International Skin Imaging Collaboration, containing 2624 clinical cases of melanoma 334, atypical nevi 144, and benign lesions 2146. The approach is compared to the prior state-of-art method on this dataset. Two-fold cross-validation is performed 20 times for evaluation 40 total experiments, and two discrimination tasks are examined: 1 melanoma vs. all non-melanoma lesions, and 2 melanoma vs. atypical lesions only. The presented approach achieves an accuracy of 93.1% 94.9% sensitivity, and 92.8% specificity for the first task, and 73.9% accuracy 73.8% sensitivity, and 74.3% specificity for the second task. In comparison, prior state-of-art ensemble modeling approaches alone yield 91.2% accuracy 93.0% sensitivity, and 91.0% specificity first the first task, and 71.5% accuracy 72.7% sensitivity, and 68.9% specificity for the second. Differences in performance were statistically significant p $$<$$ 0.05, suggesting the proposed approach is an effective improvement over prior state-of-art.

Subfields: Artificial Intelligence, Cell Biology, Oncology

View Paper: https://www.semanticscholar.org/paper/294c01931402eb5973866ae01c4ed972a5e067c2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. P. Heng
2. Qi-Yan Dou
3. Jing Qin
4. Hao Chen
5. Lequan Yu
6. LinLin Shen
7. Yuexiang Li
8. Francesco Ciompi

In Oncology

1. LinLin Shen
2. Yuexiang Li
3. P. Heng
4. Hao Chen
5. Allan Halpern
6. Noel C. F. Codella
7. Brian Helba
8. Lequan Yu

Citations in the corpus

• Attention Residual Learning for Skin Lesion Classification (2019, background): “Dermoscopy [3], [4], a recent technique of visual inspection that both magnifies the skin and eliminates surface reflection, is one of the”
• Progressive Transfer Learning and Adversarial Domain Adaptation for Cross-Domain Skin Disease Classification (2020, methodology): “Recently, deep neural networks were used to analyze pathology and clinical images and showed greatsuccess[9]–[11],includingforskindiseaserecognition[7], [12]–[17],whichalleviatesthedrawbacksofclassicalmachine learningmethods.”
• Deep features to classify skin lesions (2016, methodology): “useful for other image tasks that the CNN was not originally trained for [9, 10, 11].”
• Skin Lesion Analysis towards Melanoma Detection Using Deep Learning Network (2017, methodology): “proposed a hybrid approach, integrating convolutional neural network (CNN), sparse coding and support vector machines (SVMs) to detect melanoma [29].”
• A survey on deep learning in medical image analysis (2017, methodology): “Codella et al. (2015) Melanoma detection in dermoscopic images CNN Features from pre-trained CNN combined with”
• Skin Cancer Classification Using Convolutional Neural Networks: Systematic Review (2018, methodology): “An AlexNet model for feature extraction was also applied by Codella et al [24].”
• Deep learning ensembles for melanoma recognition in dermoscopy images (2016, methodology): “All of these features have been used in systems presented in prior literature that have achieved top performance in various medical image datasets [42], including those of dermoscopic images [32].”
• Automated Melanoma Recognition in Dermoscopy Images via Very Deep Residual Networks (2017, background): “proposed to integrate CNNs, sparse coding and support vector machine (SVM) for melanoma recognition [23].”