Title: Malignant melanoma detection using multi layer preceptron with visually imperceptible features and PCA components from MED-NODE dataset

Authors: Soumen Mukherjee; Arunabha Adhikari; Madhusudan Roy

Addresses: Department of Computer Application, RCC Institute of Information Technology, Beliaghata, Canal South Road, Kolkata 700090, India ' Department of Physics, West Bengal State University, Barasat 700126, India ' Surface Physics and Material Science Division, Saha Institute of Nuclear Physics, Sector – 1, Block – AF, Bidhannagar, Kolkata 700064, India

Abstract: In this paper, a scheme is worked out for classification of images belonging to malignant melanoma and nevus class by multi layer neural network architecture with different trainings and cost functions. Total 1,875 shape, colour and texture features are extracted from 170 images from MED-NODE dataset. With the total 1,875 features an accuracy of 82.05% is achieved. Feature ranking algorithm ReliefF is used for ranking these features. MLP is run with varying number of best ranked features. With 10 best features an accuracy of 83.33%, sensitivity of 86.77% and specificity of 72.78% are achieved with 3 fold cross-validation. Effect of pre-processing the features with principal component analysis is explored and found that the optimal number of principal components is 25, which yields a maximum accuracy of 87.18% which is much higher than the previously reported accuracy level with this dataset.

Keywords: ABCD rule; grey-level co-occurrence matrix; GLCM; grey level run length matrix; GLRLM; ReliefF; principal component analysis; PCA; autocorrelation; cross entropy; t-test.

DOI: 10.1504/IJMEI.2020.106899

International Journal of Medical Engineering and Informatics, 2020 Vol.12 No.2, pp.151 - 168

Received: 27 Dec 2017
Accepted: 28 Aug 2018

Published online: 27 Apr 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article