Computational Intelligence Paradigms in Radiological Image Processing—Recent Trends and Challenges

Abstract

Computational Intelligence (CI) refers to design, theory, application, and development of biologically and linguistically motivated computational paradigms. Traditionally the three main pillars of CI have been Neural Networks (NN), Fuzzy Systems and Evolutionary Computation. Artificial Intelligence (AI) approaches find their ways into biomedical image analysis. Radiological imaging is a crucial and regular routine in medical practice where computerized automation applications assist radiologists and clinicians to carry out regular activities within healthcare. The key advantage of radiological imaging is that health care problems can be observed directly rather than derived from symptoms. Computer-Aided Diagnosis (CADx) also known as Computer-Aided Detection (CADe) are systems that assist doctors in the interpretation of radiology imaging. In these sophisticated systems, patient's captured images are fed to computer, processed, and the final results are used to assist doctors. Such procedures require image processing and computational intelligence algorithm to process at highest speed and accuracy. To make these algorithms more intelligent, efficient and optimum, they use Fuzzy logic, Artificial Neural Network (ANN) and Evolutionary Computation. This chapter presents an overview on radiological information processing using computational intelligence paradigm, and discusses emerging trends.