In the research field of medical image analysis(MIA), machine learning(ML) will be increasingly used for more accurate and faster diagnosis, as  medical data are rapidly improving both qualitatively and quantitatively. ML methods (e.g.  CNNs, RNNs, autoencoders) incorporating mathematical and computational techniques have been proposed to extract clinically useful information  from various types of data. What is ML? Arthur Lee Samuel(1901-1990), who wrote an early TEX manual (typing system)  in 1983, defined ML as the "field of study that gives computers the ability to learn without being explicitly programmed". The core of ML is representation learning that allows a machine to learn representation of features from data set to perform a special task.

In March 2016,  ML took a big leap forward when AlphaGo, a ML program for the game Go, defeated the world's best player, Lee Sedol, in Korea. This historic match has attracted significant attention in both the scientific and popular press because of the tough computational challenges associated with playing Go proficiently.  Go has a huge number of cases, which make its complexity vastly greater than that of chess; therefore, it was regarded to be almost impossible to handle by explicit mathematical means.  However, AlphaGo seemingly handled this huge complexity without explicit programming.

 ML has the potential to deal with various ill-posed problems in MIC using statistical reasoning. Numerous mathematical models with differing integrative levels have been developed to solve various MIC problems systematically and quantitatively. However, the corresponding problems in many cases are highly nonlinear and ill posed, with modeling inaccuracies and data uncertainties, which make them difficult to deal with using solely numerical means. ML can address these issues by training a nonlinear program using existing data so as to enhance its ability to make the best predictions (or decisions) when faced with new data.