Physiological signals, medical images, and biosystems can be used to access the health of a subject and they can support clinicians by improving the diagnosis for treatment purposes. Computer-aided diagnosis (CAD) in healthcare applications can help in automated decision making, visualization and extraction of hidden complex features to aid in the clinical diagnosis. These CAD systems focus on improving the quality of patient care with a minimum of fault due to device failures. In this paper, we argue that a formal and model driven design methodology can lead to systems which meet this requirement. Modeling is not new to CAD, but modeling for systems design is less explored. Therefore, we discuss selected systems design techniques and provide a more concrete design example on computer-aided diagnosis and automated decision making.