One of the most common types of cancer in the world is skin cancer. Early detection and adequate diagnosis are crucial for effective treatment. In this paper, a novel method is proposed for automated extraction and clustering of skin lesions using Particle Swarm Optimization (PSO) algorithm. Initially, the images are preprocessed to enhance their quality. Further, a PSO-based algorithm is employed for extracting significant features from skin lesion images. The retrieved features are then used for clustering skin lesions into different groups based on their visual characteristics. The PSO algorithm is utilized to optimize the clustering process, ensuring that lesions are grouped together based on their similarities effectively. To evaluate the effectiveness of proposed method, experimentation is done on ISIC 2016 and ISIC 2017 datasets and results are compared with other approaches demonstrating the superiority of the approach.