SkinNet‐ENDO: Multiclass skin lesion recognition using deep neural network and Entropy‐Normal distribution optimization algorithm with ELM

Abstract

The early diagnosis of skin cancer through clinical methods reduces the human mortality rate. The manual screening of dermoscopic images is not an efficient procedure; therefore, researchers working in the domain of computer vision employed several algorithms to classify the skin lesion. The existing computerized methods have a few drawbacks, such as low accuracy and high computational time. Therefore, in this work, we proposed a novel deep learning and Entropy‐Normal Distribution Optimization Algorithm with extreme learning machine (NDOEM)‐based architecture for multiclass skin lesion classification. The proposed architecture consists of five fundamental steps. In the first step, two contrast enhancement techniques including hybridization of mathematical formulation and convolutional neural network are implemented prior to data augmentation. In the second step, two pre‐trained deep learning models, EfficientNetB0 and DarkNet19, are fine‐tuned and retrained through the transfer learning. In the third step, features are extracted from the fine‐tuned models and later the most discriminant features are selected based on novel Entropy‐NDOELM algorithm. The selected features are finally fused using a parallel correlation technique in the fourth step to generate the result feature vectors. Finally, the resultant features are again down‐sampled using the proposed algorithm and the resultant features are passed to the extreme learning machine (ELM) for the final classification. The simulations are conducted on three publicly available datasets as HAM10000, ISIC2018, and ISIC2019 to achieving an accuracy of 95.7%, 96.3%, and 94.8% respectively.

Multiclass skin lesion classification using deep learning networks optimal information fusion

“…Data augmentation is a typical data enhancement technique in machine learning (ML) and especially in deep learning (DL) that increases the training data [23]. Deep learning methods perform more effectively when the quantity of training data is increased [24].…”

Few‐shot learning for dermatological conditions with Lesion Area Aware Swin Transformer

“…Therefore, the causes and types of dermatological conditions are various. 1 Consequently, it's meaningful to figure out the causes and approach of accurate classification, many organizations and groups organized various competitions and build kinds of datasets looking forward to efficacious dermoscopic image analysis, like. 2 Dermatological Condition is a general term that described diseases that happen on human skin, causing red skin, acne, swelling, lesion, and other abnormal conditions.…”

“…Skin bears the brunt in the various functional activities of the body at all times as the first line of physical defense and the largest human organ, that is, skin is the very first to be attacked by a much different pathogen. Therefore, the causes and types of dermatological conditions are various 1 . Consequently, it's meaningful to figure out the causes and approach of accurate classification, many organizations and groups organized various competitions and build kinds of datasets looking forward to efficacious dermoscopic image analysis, like 2 .…”

Internet of Things-Assisted Smart Skin Cancer Detection Using Metaheuristics with Deep Learning Model

“…Dermoscopy improves melanoma diagnostic accuracy; however, it can be quite difficult to accurately analyze some cancers, and especially earlier melanomas have insufficient special dermoscopic features [11]. Although dermoscopy analyzes skin cancers with better accuracy, it is not appropriate for identifying featureless melanoma, and it requires increased accuracy to improve the patient's survival rates [12,13]. The difficulties with dermoscopy and the requirement to enhance the identification accuracy of skin tumors then positions the substructure for emerging computer-aided detection (CAD) techniques for analyzing skin cancers [14].…”