The Quantitative Evaluation of Functional Neuroimaging Experiments: Mutual Information Learning Curves

• Published in: NeuroImage (Orlando, Fla.) Vol. 15; no. 4; pp. 772 - 786
• Main Authors: Kjems, U., Hansen, L.K., Anderson, J., Frutiger, S., Muley, S., Sidtis, J., Rottenberg, D., Strother, S.C.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.04.2002
• Subjects: Adult; Brain Mapping - methods; Cerebral Cortex - physiology; cross-validation; Data Interpretation, Statistical; Female; generalization error; Humans; Image Processing, Computer-Assisted; Imaging, Three-Dimensional; learning curve; macroscopic and microscopic models; Magnetic Resonance Imaging - statistics & numerical data; Male; Mathematical Computing; Middle Aged; Models, Statistical; multisubject PET and fMRI studies; mutual information; prediction error; Psychomotor Performance - physiology; Reference Values; sensitivity map; Tomography, Emission-Computed - statistics & numerical data; cross-validation; prediction error; mutual information; multisubject PET and fMRI studies; generalization error; macroscopic and microscopic models; sensitivity map; learning curve

Learning curves are presented as an unbiased means for evaluating the performance of models for neuroimaging data analysis. The learning curve measures the predictive performance in terms of the generalization or prediction error as a function of the number of independent examples (e.g., subjects) u...