Artificial Intelligence in Skin Cancer Detection: Recent Advances and Future Directions

Skin cancer is a major global public health concern. Early diagnosis of skin cancer is essential for better patient outcomes and lower mortality rates. Artificial intelligence (AI) has the potential to increase the precision of skin cancer prediction and help doctors with diagnosis. This short communication reviews the recent advances and future directions in AI-powered skin cancer prediction. A comprehensive assessment of the literature is conducted to evaluate papers published between 2016 and 2022. The proposed research methodology explored Google Scholar, Scopus, and Web of Science databases to identify relevant studies according to proposed research questions. A total of 780 studies were found, of which 62 were from Scopus and 20 were from Web of Science These studies were selected for study as per the research question constraints. The study examined how well AI could distinguish between benign and malignant lesions and diagnose skin cancer, including melanoma and non- melanoma skin cancers. The analysis showed that deep learning algorithms showed good rates of sensitivity and specificity in predicting skin cancer, suggesting that they could be a useful tool for dermatologists. However, the small sample sizes and lack of real-world testing limited the generalizability of these findings. Additional research is needed to assess how well AI works on people with various skin tones under different environmental conditions. Future research should also focus on developing AI algorithms that can be used to diagnose other types of skin cancer, such as basal cell carcinoma and squamous cell carcinoma. The significance of this study lies in its exploration of how Artificial Intelligence (AI) can enhance skin cancer prediction. AI helps to distinguish between benign and malignant lesions, while considering limitations in sample size and real-world testing highlight which is the need for further research. The study underscores the importance of developing robust AI algorithms for diverse skin tones and various skin cancer types.