ABSTRACT
Statistical decision theory is the branch of mathematics and statistics that deals with the task of choosing among competing hypotheses based on a finite amount of data that contain some randomly varying components. In medical imaging the problem of tumor detection is an example of this kind of task. The data in this case are the output of a digital imaging device, either the raw data or the image that results from a reconstruction algorithm. The competing hypotheses are that the tumor is absent or the tumor is present in the patient. The randomness in the data has three sources, noise from the imaging system itself, anatomical and other variations in the patient population, and random variations in tumor characteristics.
