Abstract

The main purpose of the study was to develop a high accuracy system able to diagnose skin lesions using deep learning–based methods. We propose a new decision system based on multiple classifiers like neural networks and feature–based methods. Each classifier (method) gives the final decision system a certain weight, depending on the calculated accuracy, helping the system make a better decision. First, we created a neural network (NN) that can differentiate melanoma from benign nevus. The NN architecture is analyzed by evaluating it during the training process. Some biostatistic parameters, such as accuracy, specificity, sensitivity, and Dice coefficient are calculated. Then, we developed three other methods based on convolutional neural networks (CNNs). The CNNs were pre-trained using large ImageNet and Places365 databases. GoogleNet, ResNet-101, and NasNet-Large, were used in the enumeration order. CNN architectures were fine-tuned in order to distinguish the different types of skin lesions using transfer learning. The accuracies of the classifications were determined. The last proposed method uses the classical method of image object detection, more precisely, the one in which some features are extracted from the images, followed by the classification step. In this case, the classification was done by using a support vector machine. Just as in the first method, the sensitivity, specificity, Dice similarity coefficient and accuracy are determined. A comparison of the obtained results from all the methods is then done. As mentioned above, the novelty of this paper is the integration of these methods in a global fusion-based decision system that uses the results obtained by each individual method to establish the fusion weights. The results obtained by carrying out the experiments on two different free databases shows that the proposed system offers higher accuracy results.

Highlights

  • Melanoma is a type of skin cancer considered one of the deadliest forms of cutaneous cancer [1], being able to metastasize very fast

  • The other three methods that we proposed in order to detect skin lesions uses convolutional neural networks (CNNs)’s that are pre-trained with the large image databases ImageNet and

  • In this subsection we analyze the behavior of three CNNs (GoogleNet, residual networks (ResNet)-101, and NasNet-Large) as individual classifiers for melanoma detection

Read more

Summary

Introduction

Melanoma is a type of skin cancer considered one of the deadliest forms of cutaneous cancer [1], being able to metastasize very fast. In Romania, 25-30% of patients are diagnosed in advanced stages, III and IV [3]. According to the 2019 annual report of the American Cancer Society, it was estimated that there will be approximately 96,480 new cases of melanoma and 7230 people will die from the disease [4]. Melanoma usually appears as an irregular mole. Melanoma can develop on an existing mole that has changed, on a newly formed mole, but it can appear on another skin sign, or on a skin portion without any sign. More advanced lesions may display inflammation, ulceration, itching or bleeding

Objectives
Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call