MSRNet: Multiclass Skin Lesion Recognition Using Additional Residual Block Based Fine-Tuned Deep Models Information Fusion and Best Feature Selection

Abstract

Cancer is one of the leading significant causes of illness and chronic disease worldwide. Skin cancer, particularly melanoma, is becoming a severe health problem due to its rising prevalence. The considerable death rate linked with melanoma requires early detection to receive immediate and successful treatment. Lesion detection and classification are more challenging due to many forms of artifacts such as hairs, noise, and irregularity of lesion shape, color, irrelevant features, and textures. In this work, we proposed a deep-learning architecture for classifying multiclass skin cancer and melanoma detection. The proposed architecture consists of four core steps: image preprocessing, feature extraction and fusion, feature selection, and classification. A novel contrast enhancement technique is proposed based on the image luminance information. After that, two pre-trained deep models, DarkNet-53 and DensNet-201, are modified in terms of a residual block at the end and trained through transfer learning. In the learning process, the Genetic algorithm is applied to select hyperparameters. The resultant features are fused using a two-step approach named serial-harmonic mean. This step increases the accuracy of the correct classification, but some irrelevant information is also observed. Therefore, an algorithm is developed to select the best features called marine predator optimization (MPA) controlled Reyni Entropy. The selected features are finally classified using machine learning classifiers for the final classification. Two datasets, ISIC2018 and ISIC2019, have been selected for the experimental process. On these datasets, the obtained maximum accuracy of 85.4% and 98.80%, respectively. To prove the effectiveness of the proposed methods, a detailed comparison is conducted with several recent techniques and shows the proposed framework outperforms.

Keywords: classification; contrast enhancement; deep learning; feature selection; fusion; marine predator optimization; skin cancer.

Figure 1

Main flow of proposed automated melanoma recognition using deep learning.

Figure 2

A sample image of the ISIC2019 dermoscopic dataset.

Figure 3

Lesion Contrast Enhancement Results: (a,c) Original Image; (b,d) Enhanced Image.

Figure 4

Transfer learning model for the learning of deep model for skin lesion classification.

Figure 5

Original architecture of DensNet-201 CNN model.

Figure 6

Confusion matrix of Quadratic SVM after employing the proposed feature selection technique on ISIC2018 dataset. * Actinic keratosis (AK), Melanoma (MEL), Melanocytic nevus (NV), Basal cell carcinoma (BCC), Benign keratosis (BKL), Dermatofibroma (DF), and Vascular (VASC), respectively.

Figure 7

Confusion matrix of Cubic SVM after employing feature selection technique on ISIC2019 dataset.

Figure 8

Comparison of ISIC2019 dataset accuracy after employing proposed feature selection using different training/testing ratios.

Figure 9

GradCAM based visualization of fine-tuned DenseNet-201model for cancer localization.

Figure 10

Proposed method prediction in terms of Labeled images.