Chapter 3 - Deep learning applied to detection of the vulnerable atherosclerotic plaque

Abstract

Computer science has been widely adopted by modern medicine and advances in technologies and computers have enabled to become a vital tool in conventional clinical practice. Nowadays, the medical information in hospitals become larger and larger, which causes great difficulties in extracting useful information for decision support, especially when traditional manual data analysis has become inefficient and methods for computer-based analysis are indispensable. Therefore, there is a need to introduce more efficient and effective computational methods in medical analysis for decision support to help clinicians. The idea is to build decision support tools based on numerical methods that store and use knowledge from sources such as experienced clinicians, statistical analysis or computer simulations, and after that, these tools gather knowledge automatically and use optimization methods to return appropriate answers to queries or accurate predictions on future data. There are many different methods to achieve this goal; however, machine learning techniques (MLTs) have shown to be quite useful for this automation process. These techniques are based on the ability to learn from examples which is an important facet of intelligence that has been an especially fertile area of study in the last decade.