Benchmarking of deep learning algorithms for skin cancer detection based on a hybrid framework of entropy and VIKOR techniques

Baidaa Al-Bander,Hussain Mahdi,Rwayda Kh S Al-Hamd,Qahtan M Yas

doi:10.3906/elk-2103-65

Abstract

Skin cancer is one of the most common cancers worldwide caused by excessive development of skin cells. Considering the rapid growth of the use of deep learning algorithms for skin cancer detection, selecting the optimal algorithm has become crucial to determining the efficiency of computer-aided diagnosis (CAD) systems developed for the healthcare sector. However, a sufficient number of criteria and parameters must be considered when selecting an ideal deep learning algorithm. A generally accepted method for benchmarking deep learning models for skin cancer classification is unavailable in the current literature. This paper presents a multi-criteria decision-making framework for evaluating and benchmarking deep learning models for skin cancer detection based on hybridisation of entropy and VIseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) methods. Twelve well-known convolution networks are evaluated and tested on eleven publicly available image datasets to achieve the target of the study. Several criteria related to deep convolutional neural networks (CNNs) architectures, including optimisation technique, transfer learning, class balancing, transfer learning, data augmentation, and network complexity, have been considered in the multi-criteria evaluation. The decision matrix (DM) is designed based on a crossover of the five evaluation criteria and twelve (CNNs) classification models on different datasets. Subsequently, in the benchmarking and ranking of deep learning classification models, multi-criteria decision making (MCDM) techniques are used. The MCDM uses a scheme that involves the integration of entropy with VIKOR approaches. For the weight calculations of evaluation criteria, entropy is applied, while VIKOR is used to benchmark and rank the models. The obtained results reveal that the InceptionResNetV2 model gained the first rank and is selected as the optimal architecture for skin cancer detection considering the five criteria investigated in our study. The presented framework achieves a significant performance in selecting the best algorithm, which could provide substantial guidance to the researcher working in the field.

Highlights

Cancer is an irregular and uncontrolled growth of dividing cell that damages various body cells and leads to the second major cause of death in the world [1]
This section provides a description of the convolutional neural networks (CNNs) architectures and criteria considered in the study and multi-criteria decision making (MCDM) methods exploited for decision making as follows: 2.2.1
This section summarises the experiments performed to determine the efficacy of CNN models in skin cancer diagnosis

Summary

Introduction

Cancer is an irregular and uncontrolled growth of dividing cell that damages various body cells and leads to the second major cause of death in the world [1]. Automated computer-aided diagnosis (CAD) systems have become widespread among dermatologists to overcome the limitations mentioned above, reduce dermatologists’ workload, and enable rapid diagnostic rates. Deep learning (DL) has recently gained considerable popularity as one of the artificial intelligence techniques that achieved a crucial role in developing accurate and precise CAD systems due to its reliability and rapid progress. Convolutional neural network (CNN) is among the most popular algorithm used for medical image analysis [3]. There is no single deep learning classification algorithm that is superior for skin cancer diagnosis. Selecting the best DL model has posed a significant demand for decision-makers, who develop CAD systems for medical centres, in identifying and evaluating various DL classifiers for skin cancer diagnosis. Benchmarking the deep learning models under various criteria is crucial

Methods

Results

Conclusion