Comprehensive Summary
Shah et al evaluate the effectiveness and performance of Artificial Neural Network (ANN) and Convolution Neural Network (CNN) AI tools, used for early detection of skin cancer. This research utilized a literature review across various databases, including Researchgate, Google Scholar, PubMed, and IEEE Xplore. The study then discussed and evaluated ANN and CNN based on the existing research. The research found that CNN generally provides better results than ANN and other algorithms, because of its ability to classify image data more accurately. For example, one CNN model used the Adam optimiser, one of various types, and achieved a classification accuracy of 99% on a dataset using different skin cancer lesions.
Outcomes and Implications
Skin cancer is a global health concern with increasing incidence rates, therefore, there is a need for a more accurate and efficient system for skin cancer detection, as it is crucial for high survival rates and effective treatment. The study highlights the potential of using ANN and CNN to address this need. This work provides potential tools that can be used for more efficient and accurate skin cancer detection, that can lead to earlier diagnoses and improved treatment outcomes. These AI tools also offer a means to diagnose skin cancer in a more comfortable, less costly, and faster way than traditional biopsies. Although some models are said to be effective enough to be used by medical professionals currently, the authors do not provide a specific timeline for widespread clinical implementation. Instead they emphasize to the need for future research to address data standardization and scale.