Original scientific paper

https://doi.org/10.31803//tg-20210422205610

Stacked Cross Validation with Deep Features: A Hybrid Method for Skin Cancer Detection

Ahmed Al-Karawi orcid.org/0000-0002-7166-2744 ; Çukurova University, Faculty of Engineering, Electrical & Electronics Engineering Department, 01330, Balcalı, Sarıçam, Adana, Turkey
Ercan Avşar orcid.org/0000-0002-1356-2753 ; Dokuz Eylül University, Computer Engineering Department, Tınaztepe Campus, Buca 35160, Izmir, Turkey

Abstract

Detection of malignant skin lesions is important for early and accurate diagnosis of skin cancer. In this work, a hybrid method for malignant lesion detection from dermoscopy images is proposed. The method combines the feature extraction process of convolutional neural networks (CNN) with an ensemble learner called stacked cross-validation (CV). The features extracted by three different CNN architectures, namely, ResNet50, Xception, and VGG16 are used for training of four different baseline classifiers, which are support vector machines, k-nearest neighbors, artificial neural networks, and random forests. The stacked outputs of these classifiers are used to train a logistic regression model as a meta-classifier. The performance of the proposed method is compared with the baseline classifiers trained individually as well as AdaBoost classifier, another ensemble learner. Feature extraction with Xception architecture, outperforms all other benchmark models by achieving scores of 0.909, 0.896, 0.886, and 0.917 for accuracy, F1-score, sensitivity, and AUC, respectively.

Keywords

Convolutional Neural Networks; Cross Validation; Deep Learning; Dermoscopy; Skin Cancer; Stacking

Hrčak ID:

271924

URI

https://hrcak.srce.hr/271924

Publication date:

4.2.2022.

Visits: 1.267 *