In this paper is proposed a software to decision support for the detection of melanoma cancer. This approach is proposed, as this type of skin cancer is the only one that can metastasize, that is, proliferate to other organs such as lungs, liver, etc. The proposed system is elaborated through the acquisition of a set of images with 17805 samples, extractors of attributes gray level co-occurrence matrix (GLCM), Local Binary Pattern (LBP) and Central Moments. For the training and sample classification process, the Single Layer Perceptron (SLP), Multilayer Perceptron (MLP) and Support Vector Machine (SVM) classifiers are used. Subsequently, the best-evaluated model is implanted in a Raspberry Pi computer that, together with a webcam and a computer screen, allows the capture and classification of skin lesions in real-time. In the applied methodology, the authors obtained the best result of a 93 % accuracy from the use of Central Moments extractor and MLP classifier. In contrast to the state of the art of this problem, a high level of similarity is found between the accuracy rates of the Single Layer Perceptron (SLP) and the Multilayer Perceptron (MLP), demonstrating a possible resolution of the problem in a linear format.
Skin cancer is currently presented as the most common illness in the
world among fair-skinned people, the non-melanoma type being the
most common kind of skin cancer worldwide. There are two major
types of skin cancer: melanoma and non-melanoma [
Melanoma skin cancer is a malign tumor that stems from
the uncontrolled proliferation of melanocytes (pigment producing
cells)[
In this paper, therefore, is proposed a decision support software
tool that employs techniques from Computer Vision and
Artificial Intelligence using the classifiers Single Layer Perceptron (SLP)
[
The paper has the following organization: the Section 2 presents the related works that helped in the construction of this work. In Section 3 it is described the proposed method, that is, the acquisition of the set of images, elaboration of data sets, parameterization of the machine learning algorithms and implementation of the solution in a Raspberry Pi. The results obtained are presented in Session 4 and, finally, Section 5, the final considerations are presented. 2
Mariam A. Sheha, Mai S. Mabrouk and Amr Sharawy [
On another approach, H. Alquran et al [
Ansari, U. B. and Sarode, T. [
As previously mentioned, the authors of this work propose a decision support software for melanoma skin cancer detection. That said, as shown in Fig. 1, the process is divided into 6 steps. The initial 4 stages consist of implementing the computational tool, while the last two stages comprehend practical use of the software.
As shown on Fig. 1, the implementation of the tool can be divided into 4 stages: the acquisition of the image set, the elaboration of the data set, the use of machine learning techniques and the deployment on Raspberry Pi. The methodologies used for each illustrated step are presented next. 3.1
The information is extracted from an image set named ”dermoscopic pigmented skin lesions from HAM10k”3. This set contains 8903 samples of melanoma skin cancer images and 8902 samples of nonmelanoma skin cancer images (Fig. 2) , with a total of 17805 samples.
As for the ABCDE rule, professionals use The first five letters of
the alphabet a guide to help people recognize the warning signs of
melanoma. They are: Asymmetry (A) - wounds or stains are
presented asymmetrically; Borders (B) - tend to be irregular; Colors (C)
- they present different colors in the same wound; D (Diameter)
equal to or greater than 6mm; and Evolution (E) - whether it evolves
in shape, color or elevation [
In this sense, based on the aforementioned rule, the authors propose the use of attribute techniques based on textures and shapes, as they are similar to the way of detecting the problem in a practical way. The techniques used are presented next. 3 The data set is made available by Alexander Scarlat through the link: https://www.kaggle.com/drscarlat/melanoma 3.2
The authors propose the use of the extractors Gray Level
Cooccurrence Matrix (GLCM) [
D. Huang et al. [
Moments are the statistical expectation of certain power functions
of a random variable. The central moment is widely used in pattern
recognition because of their discrimination power and robustness [
Regarding the artificial intelligence algorithms, the single-layer Perceptron algorithm is implemented to verify the possibility of dealing with linear problems and, subsequently, the Multilayer Perceptron (MLP) and the Support Vector Machine, as they present themselves as robust algorithms to obtain better results, in case of a nonlinear problem.
To verify the efficiency of the proposed methodology, experiments are carried out with the Single Layer Perceptron (SLP) classifier, using learning rates of 0.01, 0.05, 0.1 and 0.5, signal function and reduction linear or exponential learning rates.
For the MLP classifier, we implemented several learning rates that vary between 0.1 and 0.5, linear or exponential reduction of these rates, logistic function or hyperbolic tangent. The values referring to the number of neurons in the hidden layer of the MLP are entered empirically, ranging from 2 to 100 neurons.
For the SVM classifier, 4 types of kernels are used; linear, RBF, polynomial and sigmoid. The value of C, that is, the penalty parameter is entered from the following values 0.01, 0.1, 1.0 and 10.
For both algorithms, the training is conditioned to end when it reaches an error rate of 10 5 and there are 2000 iterations, in which the weights with the best accuracy are chosen as the ones that best represent the results of the training.
As mentioned, the authors used 17.805 images in total, which are divided into 8.903 samples of the class ”melanoma” and 8.902, of ”non-melanoma”. In this sense, 7.122 samples selected at random for each type of class were used to compose the training/tests set. The rest of the data, that is, 1.781 samples of the class ”melanoma” and 1.780 samples of the class ”non-melanoma”, make up the test set.
For all classifiers, the K-Fold cross-validation technique is used, where a value of K equals 10. The application of this method consists of dividing the total set of training data into K subsets of the same size (Fig. 3). Thus, each subset is used for validation, while the rest of the set is applied to estimate the parameters.
As shown in Fig. 3, the data set is divided into 2 parts. The first is divided into 10 subsets that are used in the training and validation process. Finally, the second part of the set, unknown by the classifier, is used for the test process. 3.4
In order to make this application usable to healthcare professionals as well as to attain low acquisition costs, the computer program (Fig. 4) was deployed to a Raspberry Pi 3B+ computer.
After installing the classification software on the raspberry computer, we can use the webcam to obtain images of the skin for analysis, and the computer screen for verifying the suggested results. 3.5
As previously mentioned, the authors propose a decision support software. Thus, the results generated by it must be analyzed by a doctor for proper conclusions.
In the Table 1 shows the best results obtained in this study. This solution is obtained with the Central Moments extractor and Multilayer Perceptron classifier. It should be noted that, in order to obtain the best results, the authors use various combinations between their attributes generated by the extractors. Thus, were obtained the best results (Table 1) by applying the following Central Moments: mu11, mu21, mu12, mu30, mu03.
Although the Multilayer Perceptron classifier offers a better solution from the data obtained by the Central Moments extractor, the authors perceive similar results from the same attributes used with the Single Layer Perceptron classifier. This perception demonstrates that the problem of detecting melanoma cancer can be linearly separable, making it an easily resolvable problem. Table 2 illustrates the best results obtained from the classifier SLP.
It is important to emphasize that by using the Central Moments extractor, one can divide the data almost linearly. Therefore, the decision limits for both classifiers have similarities and simplicity in their structures. 5
In view of the literature involved, this work was differentiated by using less robust techniques and achieving similar or even superior results. Providing a lower computational cost in the early detection of melanoma cancer.
Although most Artificial Intelligence models require powerful processing and extensive memory resources, there are still methods that, when applied correctly, produce satisfactory results. Therefore, the exact application of an attribute extractor related to the binary detection of melanoma skin cancer produces a good solution to the problem without requiring much processing power or memory. To that extent, the use of attribute extractors suitable for a problem can simplify its classification. In this scenario, a Perceptron can be a significant resource for achieving satisfactory results.
It should be noted that when deploying the software on the raspberry pi board, a simple, fast and portable solution is created, which enables support the decision to health care professionals regarding melanoma skin cancer detection. The software also helps improve accessibility to patients who live in regions of difficult access, since the device is portable enough for professionals to provide home care when the patient is unable to reach the care unit.