The Method and System for Evaluating the Quality of Medical Image dataset for Machine Learning

摘要

The present invention relates to a method for evaluating the quality of a medical image data set to check whether the medical image data is suitable for machine learning, and a system therefor. Evaluation criteria can include: data normality, which means a ratio of normal frames among the entire frames; learning suitability, which means a ratio of labeled or labelable frames among received data; and anatomical completeness, which means a ratio of anatomical elements included in the received data compared to anatomical elements based on a medical standard.