Multilevel Threshold Segmentation of Skin Lesions in Color Images Using Coronavirus Optimization Algorithm

Area of Science:

• Medical Imaging
• Computational Intelligence
• Dermatology

Background:

• Skin cancer (SC) poses a significant health risk due to its high mortality rate.
• Early detection of skin cancer is crucial for effective treatment outcomes.
• Multilevel Thresholding (MLT) is a key technique for segmenting regions of interest in medical images.

Purpose of the Study:

• To develop and evaluate a novel meta-heuristic algorithm for improved skin cancer image segmentation.
• To address the Multilevel Thresholding (MLT) challenges in segmenting skin cancer images.
• To enhance the early detection capabilities for skin cancer through advanced image analysis.

Main Methods:

• Utilized the Coronavirus Disease Optimization Algorithm (COVIDOA) for Multilevel Thresholding (MLT) of skin cancer images.
• Integrated Otsu, Kapur, and Tsallis methods as fitness functions within the COVIDOA framework.
• Compared the proposed COVIDOA against Arithmetic Optimization Algorithm (AOA), Sine Cosine Algorithm (SCA), Reptile Search Algorithm (RSA), Flower Pollination Algorithm (FPA), Seagull Optimization Algorithm (SOA), and Artificial Gorilla Troops Optimizer (GTO).

Main Results:

• The proposed COVIDOA algorithm demonstrated superior performance in segmenting skin cancer images compared to existing meta-heuristic algorithms.
• Achieved lower Mean Square Error (MSE) and higher Peak Signal-To-Noise Ratio (PSNR), indicating improved image segmentation quality.
• Showcased enhanced Feature Similarity Index Metric (FSIM) and Normalized Correlation Coefficient (NCC) values, confirming segmentation accuracy.

Conclusions:

• The hybrid COVIDOA approach effectively solves the Multilevel Thresholding (MLT) problem for skin cancer image segmentation.
• The algorithm's superior performance in key segmentation metrics (MSE, PSNR, FSIM, NCC) supports its clinical utility.
• This advancement in image segmentation technology holds promise for improving early skin cancer detection and patient outcomes.