Dermatology diagnosis with feature selection methods and artificial neural network

Abdul-Rahman, Shuzlina; Norhan, Ahmad Khairil; Yusoff, Marina; Mohamed, Azlinah; Mutalib, Sofianita

doi:10.1109/iecbes.2012.6498195

Cited by 14 publications

(9 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A feed forward back propagation neural network (FFBP-NN) is used in [21,33] for dermatological disease diagnosis. ANNs have also been used in other skin disorder classification applications: [34,36,38]. MLP has been used in [47] for cardiovascular disease diagnosis.…”

Section: Classification Methodsmentioning

confidence: 99%

“…Other skin disorders can be detected by image processing applications that verify the existence of a single disease e.g., psoriasis in [29] and acne in [30,31]. Other applications discriminate between multiple skin diseases [32][33][34][35][36][37][38]. The image processing in most of these cases is performed in RGB although different color spaces have also been employed like hue saturation value (HSV), YCbCr [30].…”

Section: Introductionmentioning

confidence: 99%

“…Smart phone applications are also available for skin disorder diagnosis like those described in [31,39,40]. The dataset sources in these references are Atlas of Clinical Dermatology (Churchill Livingstone, 2002), eCureMe Online Medical Dictionary [33], UCI Repository of Machine learning databases [34,38], Dermnet Skin Disease Atlas [36], in the references of [37].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Review of Image Processing Techniques Common in Human and Plant Disease Diagnosis

Petrellis

2018

Symmetry

View full text Add to dashboard Cite

Image processing has been extensively used in various (human, animal, plant) disease diagnosis approaches, assisting experts to select the right treatment. It has been applied to both images captured from cameras of visible light and from equipment that captures information in invisible wavelengths (magnetic/ultrasonic sensors, microscopes, etc.). In most of the referenced diagnosis applications, the image is enhanced by various filtering methods and segmentation follows isolating the regions of interest. Classification of the input image is performed at the final stage. The disease diagnosis approaches based on these steps and the common methods are described. The features extracted from a plant/skin disease diagnosis framework developed by the author are used here to demonstrate various techniques adopted in the literature. The various metrics along with the available experimental conditions and results presented in the referenced approaches are also discussed. The accuracy achieved in the diagnosis methods that are based on image processing is often higher than 90%. The motivation for this review is to highlight the most common and efficient methods that have been employed in various disease diagnosis approaches and suggest how they can be used in similar or different applications.

show abstract

Section: Classification Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Review of Image Processing Techniques Common in Human and Plant Disease Diagnosis

Petrellis

2018

Symmetry

View full text Add to dashboard Cite

show abstract

“…Feed forward back propagation artificial neural network is used in most of the existing systems, where the nonlinear data may not be evenly distributed. The data gets transformed by the Hidden layer (middle layer) of the multilayer network [3]and learns the data transformation [4] to make it linearly separable. This is an approach which gives comparatively less accurate results…”

Section: Existing Systemmentioning

confidence: 99%

Convolutional Neural Network Model for Predicting Skin Based Diseases and Evaluation of Risk Assessment

Jainulabudeen¹,

Murugavalli²,

Shalma³

2019

IJRTE

View full text Add to dashboard Cite

In today's dynamic lifestyle, with people not prioritizing hygiene as an essential entity, they tend to get more easily prone to skin diseases. As a result, it has become highly significant to devise an automated mechanism which helps users to predict a disease using simple methodologies such as input in the form of images. In our case we have used Convolutional Neural Networks to diagnose the same. The disease prediction system uses simple methodologies including taking user input in the form of images, which aids in providing more accurate results. As the concept used is Convolutional Neural Network (CNN), the system not just analyses the input and predicts the disease based on it, but also guesses the nearest possible result based on its feature and adds that record to that class of disease for future predictions. Over a longer period of time, this majorly helps in preventing hazardous skin diseases from causing widespread damage, as spreading of skin diseases is highly rapid and difficult to control, the ultimate aim through this project is to overcome the above limitations and help create awareness amongst people onskin diseases, which happens to be a growing concern in the near future.

show abstract

“…FCBFFast Correlation-Based Filter method is applied in the study for the prediction of Type-II diabetes [14]. To get the greater classification accuracy and minimum response time cfs and FCBF methods are applied over dermatology dataset [15]. Using those methods, a minimum subset of features are collected and classification model is generated with those features.…”

Section: Introductionmentioning

confidence: 99%

A Novel M-Cluster of Feature Selection Approach Based on Symmetrical Uncertainty for Increasing Classification Accuracy of Medical Datasets

Potharaju¹,

Sreedevi²

2017

JESTR

View full text Add to dashboard Cite

In recent days, due to the advancements in technology, a massive amount of data is generating in every area of study, including the medical field. This massive amount of data contains a large number of attributes and instances in it. It is not an easy task for classification and prediction from this high dimensional data. Because, all the attributes in the dataset can't give an impressive result in classification and prediction. Now, it is unavoidable to reduce the high dimensional data for better classification result, which is possible by feature selection and reduction techniques .In this research paper, a novel M-Cluster feature selection (Mcfs) based on Symmetrical Uncertainty (SU) Attribute Evaluator is proposed for improving the classification accuracy of medical datasets. The proposed approach divides the total feature space into 'M' clusters, each cluster has a finite set of attributes in it without any duplication. Feature subset formed by proposed technique is tested using Dermatology and Breast Cancer medical datasets, and compared with an existing filter-based feature selection techniques(Information Gain (IG), Chi-Squared (Chi), Gain Ratio Attribute Evaluator (GR), ReliefF (Rel) ).Experimental results displayed an improved performance with some of the clusters formed by proposed method than existing methods. For experimenting proposed technique, KNN-Lazy learner, Naive Bayes (NB) Classifier, J48-Rule based learner, JRip -Tree based learners are used.

show abstract

Dermatology diagnosis with feature selection methods and artificial neural network

Cited by 14 publications

References 7 publications

A Review of Image Processing Techniques Common in Human and Plant Disease Diagnosis

A Review of Image Processing Techniques Common in Human and Plant Disease Diagnosis

Convolutional Neural Network Model for Predicting Skin Based Diseases and Evaluation of Risk Assessment

A Novel M-Cluster of Feature Selection Approach Based on Symmetrical Uncertainty for Increasing Classification Accuracy of Medical Datasets

Contact Info

Product

Resources

About