Augmented Intelligence enabled Deep Neural Networking (AuDNN) framework for skin cancer classification and prediction using multi-dimensional datasets on industrial IoT standards

Abstract

Skin Cancer, the most common cancer in developing nations with unprecedented diagnosis and prediction. The disease has recorded 500,000 new cases in United States and placed as 19th most common cancer globally. In this article, a novel approach of skin cancer classification and prediction using augmented intelligence is proposed. The technique is processed on Kaggle Re-Snet50 datasets appended on deep neural networking framework. The Re-Snet50 is 8x-deeper than VGG net datasets. The Augmented Deep Neural Networking (AuDNN) technique has extracted arbitrary features with RoI identification for cancer region extraction and clustering. The extracted datasets are synchronized with multilayer attribute dependency mapping for improved prediction ratio. The proposed technique is developed on the terminology of Industrial IoT standards to assure a reliable communication framework for effective communication and coordination on global instrumental networking ecosystem via dual cross reference validation technique. The technique has clearly outperformed the previous approaches due to augmented intelligence based DNN multi-dimensional mapping. The technique has achieved 93.26% accuracy in classification and prediction of skin cancer on rational computation.