SKIN DISEASE CLASSIFICATION BASED ON HYBRID WAVELET FEATURES AND ENSEMBLE METHODS

Abstract

Skin cancer is one of the most common cancers in the world. However, the disease is curable if detected in the beginning stage. Early detection of malignant lesions through accurate techniques and innovative technologies has a significant impact on reducing skin cancer mortality rates. Recently, artificial intelligence has come to the forefront to facilitate skin cancer diagnosis based on medical images. This work is concerned mainly with the analysis of dermoscopic images from PH2 database, the extraction of quantitative parameters that may describe the pattern for Skin Cancer Diagnostic System. Textural information is the major concern of the study. To show how textural patterns can be well described by texture specification, co-occurrence wavelet features are extracted. Then, Random Forest (RF) with ensemble method-based classifier is used for classification. Accuracy, Precision, Sensitivity, f1-score are considered as a measure of the usefulness of theproposed model.

Imprint

V. Vidya Lakshmi, S.J. Grace Shoba, Angelina Royappa, J. Gnana Arun Johnson SKIN DISEASE CLASSIFICATION BASED ON HYBRID WAVELET FEATURES AND ENSEMBLE METHODS. Cardiometry; No.26 February 2023; p.-; DOI: .; Available from: https://www.cardiometry.net/issues/no26-february-2023/skin-disease-classification-based-on-hybrid-wavelet-features