Design and Evaluation of a Multi-model, Multi-level Artificial Neural Network for Eczema Skin Lesion Detection

Abstract

There are several current systems developed to identify common skin lesions such as eczema that utilize image processing and most of these apply feature extraction techniques and machine learning algorithms. These systems extract the features from pre-processed images and use them for identifying the skin lesions through machine learning as the core. This paper presents the design and evaluation of a system that implements a multi-model, multi-level system using the Artificial Neural Network (ANN) architecture for eczema detection. In this work, multi-model system is defined as architecture with different models depending on the input characteristic. The outputs of these models are integrated by a decision layer, thus multi-level, which computes the probability of an eczema case. The resulting system has 68.37% average confidence level as opposed to the 63.01% of the single level, i.e. single model, system in the actual testing of eczema versus non-eczema cases. Furthermore, the multi-model, multi-level design produces more stable models in the training phase wherein overfitting was reduced.

Use of artificial intelligence in dermatology

“…To improve the accuracy of current research, contextual information could be added for consideration by AI programs. [ 4 23 ]…”

Machine Learning in Dermatology: Current Applications, Opportunities, and Limitations

“…This is a notable limitation considering that the incidence of scalp psoriasis is 45-56% and nail psoriasis is 23-27% among psoriatic patients [57]. CNNs have been created for other diseases, such as atopic dermatitis [58], onychomycosis [56], and rosacea [59]. To classify onychomycosis, Han et al [56] used a R-CNN to generate a training datasets of 49,567 images of nails and found that a combination of their datasets performed better than dermatologists.…”

Artificial intelligence in dermatology and healthcare: An overview

“…105 Automated diagnosis of other erythemato-squamous diseases such as seborrheic dermatitis, atopic dermatitis, lichen planus, pityriasis rosea and pityriasis rubra pilaris has been studied using various clinical and histopathological features. 106,107 Acne New technologies in imaging and software solutions have been developed in acne and rosacea evaluation. A study by Min et al showed that compared with manual counting performed by an expert dermatologist, the sensitivity and positive predictive value of the lesion-counting program was greater than 70% for papules, nodules, pustules, and whitehead comedones.…”