AN IMPROVED DATA COLLECTION FRAMEWORK FOR CONTINUOUS LEARNING MECHANISM IN ML OR DL BASED MEDICAL IMAGING SYSTEMS
Publication Date: 2019-Oct-31
The IP.com Prior Art Database
The present disclosure proposes an improved data collection framework for continuous learning mechanism in Machine Learning (ML) or Deep Learning (DL) based medical imaging systems. The multiple users span across the hospital ecosystem using an interactive UI approach. The framework for continuous learning is obtained by involving various experts or users of the ecosystem across the breadth and depth of the healthcare ecosystem. The data collection and segregation are implemented to improve the quality of ML or DL prediction in the health care domain. The interactive user interface is utilized to collect data collection and continuously update the database of the medical imaging system. The data is assessed at the exam level and extended further at a granular level based on the requirements of the system. The quality of the data is enhanced both in-depth and breadth in a dynamic manner by multiple users spanning across the hospital ecosystem using an interactive UI approach.