A Smartphone-Based Skin Disease Classification Using MobileNet CNN

TL;DR

This paper presents a mobile application for skin disease classification using MobileNet CNN, achieving high accuracy through various data balancing and preprocessing techniques, and deploying the model on Android.

Contribution

It introduces a skin disease classification system on Android utilizing transfer learning with MobileNet and explores data balancing methods to improve accuracy.

Findings

Achieved 94.4% accuracy with oversampling and data augmentation.

Using imbalanced data with default preprocessing yielded 93.6% accuracy.

The model was successfully deployed on an Android application.

Abstract

The MobileNet model was used by applying transfer learning on the 7 skin diseases to create a skin disease classification system on Android application. The proponents gathered a total of 3,406 images and it is considered as imbalanced dataset because of the unequal number of images on its classes. Using different sampling method and preprocessing of input data was explored to further improved the accuracy of the MobileNet. Using under-sampling method and the default preprocessing of input data achieved an 84.28% accuracy. While, using imbalanced dataset and default preprocessing of input data achieved a 93.6% accuracy. Then, researchers explored oversampling the dataset and the model attained a 91.8% accuracy. Lastly, by using oversampling technique and data augmentation on preprocessing the input data provide a 94.4% accuracy and this model was deployed on the developed Android application.