Current Status of Computational Intelligence Applications in Dermatological Clinical Practice

1. INTRODUCTION

Intelligent computer systems already provide support to healthcare professionals.

Computational intelligence has been widely researched and applied owing to its ability to cope with large amounts of clinical data and uncertain information [1]. Computational intelligence uses algorithms based on biological data, primarily on neuronal functioning. The three pillars on which computational intelligence is based are neural networks, genetic algorithms, and fuzzy systems [2]. Neural networks are algorithms employed for function approximation or classification problems. They include supervised, unsupervised, and reinforcement learning [3].

Constant progress is being made on computational intelligence to improve the prevention, diagnosis, and treatment of diseases in general, and mycosis in particular [4].

The yeast Candida albicans is the most prevalent pathogenic Candida species (spp.), although the increase in immunocompromised patients has been accompanied by an increase in the diversity of pathogenic strains found as etiological agents of fungal infections [5, 6].

The most typically used method to identify Candida spp. is the chromogenic culture medium. An analysis of the growth facilitates the determination of the purity of the colonies, and the identification of Candida spp [7].

The samples used by us are from this culture medium and were later imaged under an optical microscope. After reviewing several images, we have selected C. albicans, C. parapsilosis, C. krusei, and C. kefyr for our project. The images used in the elaboration of the program correspond to Figs. (1-7).

2. MATERIALS AND METHODS

2.2. Description of the Method

Development of programs for the detection of Candida species

As the characteristics of the different species of Candida are known, we applied digital image processing with Octave. We performed an adaptation of the diagnostic methods of the Candida and their subsequent study under a microscope for the classification of the different species Most of these methods have typical concepts, such as the variety of colors, asymmetry in their internal forms, and abrupt and irregular edges, and are used to identify elements of the said fungi under the optical microscope. These are the primary concepts that are analyzed in the programs performed, and are referred to as the “Candida Identifier.”

4.3. To Determine Our Model Behavior When we Apply it to Candida sp., we Constructed a Confusion Matrix (Table 2 &3)

Table 2.

Comparison	Identifier Value
Real Value	x/x	C. albicans	C. krusei	C. kefyr	C. parapsilosis
	C. albicans	100	0	0	0
	C. krusei	0	90	10	0
	C. kefyr	0	10	90	0
	C. parapsilosis	0	0	0	100

Table 3.

Hit rate = Identification / Actual value Hits C. albicans:100100, 100% Hits C. krusei: 90100, 90% Hits C. Kefyr: 90100, 90% Hits C. parapsilosis: 100100, 100% The images used in the elaboration of the program correspond to figures 1, 2, 3, and 4.

4.5. Interface from Database of Java Script Libraries

Finally, as Octave does not have a method to generate the GUI, the interface has been implemented through the Electron framework, which is developed in JavaScript, and HTML (Table 5).

Table 4.

function [identificador] = leeImagen(file) pkg image load; % This part reads the file of the image that will be classified by the% recognition program. RGB = imread (file); % The RGB image is passed in grayscale, that is, the original image with its component% Red, Green, and Blue I = rgb2gray (RGB); % In this stage, the image that is in ray scale is binarized. This operation passes% first from the format uint8 to double. Having this format, the scale of gray% is compared with the threshold 0.2. Values above 0.2 become! (white) and below% 0 (black). BW = im2bw(I,0.2); % The image is displayed % imshow (BW) % The bwlabel command labels the binary image, such that it establishes contiguous (neighboring) zones to determine the regions. abeledImage = bwlabel (BW); % The regionprops function allows for the determination of the geometric descriptors of the figures obtained, after binarizing the images. The areas are found, as well as the perimeters that are essential% to initially identify the circular regions. This descriptor allows the% fungi types to be differentiated. Thus, for example, the circular area of a Candida% albicans fungus differentiates it from a parapsilosis fungus%. measurements = regionprops (labeledImage, 'Area', 'Perimeter'); allAreas = [measurements.Area]; allPerims = [measurements.Perimeter]; circularities = allPerims. ^ 2./ (4 * pi * allAreas); blobMeasurements = regionprops (labeledImage, 'BoundingBox', 'Area'); allBlobAreas = [blobMeasurements.Area]; % Find objects that have “round” values of circularities. roundObjects = find (circularities <4); % Whatever value you want. % Compute new binary image with only the round objects in it. binaryImage = ismember (labeledImage, roundObjects)> 0; identificador = find(circularities < 4 & allBlobAreas > 10000); %figure(2) %imshow(binaryImage); Endfunction % In this stage, the images that are being loaded to the application% are simply classified according to the identifier that has been defined for the fungi in the function. % clear all; clc; # not a function file: pkg image load; file = 'ima.jpg'; [identifier] = readImage (file); if length (identifier) == 1 && (identifier <10) disp (“The fungus is Candida parapsilosis”) elseif length (identifier) == 1 && (identifier> 20) && (identifier <30) disp (“The fungus is Candida albicans”) elseif length (identifier> 100) disp (“The fungus is Candida kefyR”) else disp (“The fungus is Candida krusei”) endif fflush.stdout ()

Table 5.

<! DOCTYPE html> <html> <head> <meta charset = “UTF-8”> <title> Mushroom Recognition </ title> </ head> <style> .body { background: # 000; color: #fff; overflow: hidden; } </style> <body class=”body”> <input type=”file” onchange=”onImageSelected(event)”/> <button type=”button” id=”htmlButtonPress” onclick=”launchPython()” >Clasifica</button> <img id=”myimage” width=600 height=400 > <canvas id=”myCanvas”/> <section> <script> var buffer = require('buffer'); var path = require('path'); var fs = require('fs'); // Here, the image that is identified in the Canvas is loaded as “myimage. function onImageSelected(event) { var selectedFile = event.target.files[0]; var reader = new FileReader(); var imgtag = document.getElementById(“myimage”); reader.onload = function(event) { imgtag.src = event.target.result; var imagen = imgtag.src; var data = imagen.replace(/^data:image\/\w+;base64,/, “"); // console.log(“La data es:”+data) var buf = new Buffer(data, 'base64'); fs.writeFile('ima.jpg', buf, (error) => {/* handle error*/}); }; reader.readAsDataURL(selectedFile); } const PowerShell = require(“powershell”); // Here, the communication with Octave is made so that it executes the programs. function launchPython() { let ps = new PowerShell('C:/Octave/octave-4.4.1/bin/octave-cli -qf demos.m'); ps.on(“error”, err => { console.error(err); }); // Stdout ps.on(“output”, data => { console.log(data); const Par = document.createElement('p') Par.textContent = data; document.querySelector('body').appendChild(Par); }); } </script> // Modules to control application life and create native browser window const {app, BrowserWindow} = require('electron') // Keep a global reference of the window object. If you do not, the window will // be closed automatically when the JavaScript object is garbage collected. let mainWindow function createWindow () { // Create the browser window. mainWindow = new BrowserWindow({width: 600, height: 700}) // and load the index.html of the app. mainWindow.loadFile('index.html') // Open the DevTools. // mainWindow.webContents.openDevTools() // Emitted when the window is closed. mainWindow.on('closed', function () { // Dereference the window object, typically you would store windows // in an array if your app supports multi windows, this is the time // when you should delete the corresponding element. mainWindow = null }) } // This method will be called when Electron has finished // initialization and is ready to create browser windows. // Some APIs can only be used after this event occurs. app.on('ready', createWindow) // Quit when all windows are closed. app.on('window-all-closed', function () { // On OS X, it is typical for applications and their menu bar // to stay active until the user quits explicitly with Cmd + Q if (process.platform!== 'darwin') { app.quit() } }) app.on('activate', function () { // On OS X, it is typical to re-create a window in the app when the // dock icon is clicked and there are no other windows open. if (mainWindow === null) { createWindow() } }) // In this file, you can include the remainder of your app's specific primary process // code. You can also place them in separate files and require them here. See translate video of program operation in Appendix 1.

5. DISCUSSION

A major difficulty was in obtaining a large sample and images of the Candida spp. used with similar characteristics, because the time of planting and the technique vary according to its shape and size. The system cannot change the size of the images during the entire process [9].

Another difficulty that also influences the classification of the image is the image quality [10, 11].

The more the noise or the poorer the image quality, the less precision the edge mapping process will have [12].

Further, the less accuracy on the border mapping process, the worse is the performance on the classification process [13].

In the literature, we did not find models that were used to recognize Candida spp. However, we found references bout the artificial neural networks to calculate antifungal activity against C. albicans. Furthermore, there are references to other cutaneous diseases [14-16] such as melanoma [17].

Hence, we established a goal. The objective was to reach a diagnostic accuracy of at least 80%. The model proposed by Manousaki et al. [18] provided an accuracy of 89.4%. The model for pre-diagnostic digital imaging developed by Christensen et al. [19] presented an accuracy of 77%. Our model is superior in accuracy to those mentioned above.

The final prediction model developed exhibits an accuracy of almost 90% and will thereby represent an improvement in diagnostic accuracy. Despite the different appearances and qualities of the images in the image database, we could develop a prediction model with accuracy better than that expected from an experienced dermatologist [20].

The algorithms are thereby stable and not dependent on the quality of the image |under inspection. These new tools can also be utilized in telemedicine, where images can be uploaded to an automated web-based database and subsequently analyzed [21].

CONCLUSION

We designed and programmed the “Candida identifier application” in Octave. It allowed the identification of the Candida species by locating certain geometric descriptors, such as the centroid and the surfaces of the circular objects that comprise the images. This system could differentiate the C. albicans from other varieties of Candida such as C. Kefyr, C. parapsilosis, and C. krusei, with accuracy. The analysis by the system of these descriptors allowed us to identify the Candida species in more than 90% of the cases. The sensitivity and specificity that were above 90% indicated a high diagnostic efficacy that was rather large to be the first version of a diagnostic program.

The application was highly precise in the identification based on the forms; however, it did not exhibit the same features in color-based identification. This element typically presents variations in the data, thus rendering identification difficult. We should acquire more copious databases to implement more precise identification methods, such as deep learning algorithms for images; however, the recognition of information in three different areas was satisfactory, as it allowed us to determine and classify injuries within the medical margins.

Finally, we emphasize that the objectives set at the beginning of the project were achieved. The results that we obtained from the “Candida identifier system” paved the way for handling images obtained from an optical microscope. This implies a wider range of materials can be used and specialized personnel economy (mainly in developing countries), and better service from the patient's perspective.

REFERENCES