Application of the mutual information criterion for feature selection in computer-aided diagnosis

• Published in: Medical physics (Lancaster) 2001-12, Vol.28 (12), p.2394-2402
• Main Authors: Tourassi, Georgia D., Frederick, Erik D., Markey, Mia K., Floyd, Carey E.
• Format: Article
• Language: English
• Subjects: acute pulmonary embolism; Computational complexity; Computer aided diagnosis; computer‐assisted diagnosis; Data analysis; Diagnosis, Computer-Assisted - methods; feature extraction; feature selection; genetic algorithms; Genomic techniques; Humans; Image analysis; image texture; Information and communication theory; lung; Lungs; medical image processing; Models, Statistical; mutual information; Normal Distribution; Physicists; Software
• ISSN: 0094-2405; 2473-4209
• DOI: 10.1118/1.1418724

The purpose of this study was to investigate an information theoretic approach to feature selection for computer-aided diagnosis (CAD). The approach is based on the mutual information (MI) concept. MI measures the general dependence of random variables without making any assumptions about the nature of their underlying relationships. Consequently, MI can potentially offer some advantages over feature selection techniques that focus only on the linear relationships of variables. This study was based on a database of statistical texture features extracted from perfusion lung scans. The ultimate goal was to select the optimal subset of features for the computer-aided diagnosis of acute pulmonary embolism (PE). Initially, the study addressed issues regarding the approximation of MI in a limited dataset as it is often the case in CAD applications. The MI selected features were compared to those features selected using stepwise linear discriminant analysis and genetic algorithms for the same PE database. Linear and nonlinear decision models were implemented to merge the selected features into a final diagnosis. Results showed that the MI is an effective feature selection criterion for nonlinear CAD models overcoming some of the well-known limitations and computational complexities of other popular feature selection techniques in the field.