An Improved Skin Lesion Boundary Estimation for Enhanced-Intensity Images Using Hybrid Metaheuristics

Abstract

The demand for the accurate and timely identification of melanoma as a major skin cancer type is increasing daily. Due to the advent of modern tools and computer vision techniques, it has become easier to perform analysis. Skin cancer classification and segmentation techniques require clear lesions segregated from the background for efficient results. Many studies resolve the matter partly. However, there exists plenty of room for new research in this field. Recently, many algorithms have been presented to preprocess skin lesions, aiding the segmentation algorithms to generate efficient outcomes. Nature-inspired algorithms and metaheuristics help to estimate the optimal parameter set in the search space. This research article proposes a hybrid metaheuristic preprocessor, BA-ABC, to improve the quality of images by enhancing their contrast and preserving the brightness. The statistical transformation function, which helps to improve the contrast, is based on a parameter set estimated through the proposed hybrid metaheuristic model for every image in the dataset. For experimentation purposes, we have utilised three publicly available datasets, ISIC-2016, 2017 and 2018. The efficacy of the presented model is validated through some state-of-the-art segmentation algorithms. The visual outcomes of the boundary estimation algorithms and performance matrix validate that the proposed model performs well. The proposed model improves the dice coefficient to 94.6% in the results.

An efficient decision support system for leukemia identification utilizing nature-inspired deep feature optimization

“…First, a population matrix P consisting of N p row vectors was generated, where each vector was composed of four variables, i.e., α , β , γ , and δ . Each entity of the population matrix was generated randomly, as Equations 3 – 6 ( 43 ).…”

“…In the next step, each population vector P i i ∈{1, …, N p } underwent mutation operation to generate its corresponding mutation vector M i such that ( 43 )…”

“…The population vector P i and its corresponding mutation vector M i then underwent binomial crossover operation to generate vector C i such that ( 43 )…”

“…Next, all decision variables each vector C i ( t ) are bounded within limits l b and u b . In differential evolution, a greedy selection of survival of the fittest was carried out using the following criteria to update the population matrix ( 43 ).…”

Multiclass skin lesion classification using deep learning networks optimal information fusion

“…The chance of a higher sample class increases when the dataset object classes are unequal. The morphology, color, and texture of multiclass skin lesion types are identical, and they share common features [22]. Those features are later classified into an incorrect skin class.…”