Using Edge-based Deep Learning Model for Early Detection of Cancer

Cancer is one of the most frequent causes of death in the world. Usually, cancer can be easily diagnosed if characteristic symptoms occur. However, many people who are suffering from cancer have no symptoms. Early diagnosis of tumors is essential to contrast their progression, helping to define more effective treatments to provide long-term survival. Early cancer detection is effective if sensible data can be investigated through high-performance technologies like edge computing. Edge computing is a new paradigm for analyzing data as close to the source as possible, avoiding exporting them outside. Hence, edge-based deep learning models can be applied to improve early cancer detection. This paper provides an use case of a classification task on tumor-related data based on the famous UCI machine learning data sets repository using a deep learning approach based on edge computing. In addition, the manuscript provides an overview of the edge computing paradigm, highlighting its advantages and usability. We also described a small experiment with real tumor data to characterize performance considerations. Moreover, the presented model can be used with different data types, such as images, EGC, and ECC signals.