Abstract:
In the last decade, deep learning has been transforming the healthcare scenario with the provision of new computer-assisted diagnosis tools in an unprecedented way. The s...View moreMetadata
Abstract:
In the last decade, deep learning has been transforming the healthcare scenario with the provision of new computer-assisted diagnosis tools in an unprecedented way. The spread of specialized hardware and software has been supporting this reality. The development of such predictive models, however, requires expertise and time. In this context, this article proposes and describes the implementation of a no-coding software framework for the automation of deep learning training processes, aiming to support the development of medical image diagnostic classifiers by healthcare professionals with limited expertise in deep learning. It is a web solution that allows the easy management of data, models, and training processes associated with user requests, where the benefits extend to non-experts in the machine learning field. It is an intuitive web solution that contains a set of additional modern tools like Automatic Machine Learning (AutoML) to aid design models and a ranking system to help improve and share them.
Date of Conference: 09-12 December 2021
Date Added to IEEE Xplore: 14 January 2022
ISBN Information: