Improved small volume lung cancer detection with computer-aided detection: database characteristics and imaging of response to breast cancer risk reduction strategies

Computer-aided detection (CAD) and diagnosis (CADx) of in vivo imaging studies are important tools based on bioinformatics. Currently, there are two diseases for which the United States Food and Drug Administration (FDA) has given premarket approval (PMA): the detection of signs consistent with lung cancer on chest radiographs and breast cancer on mammograms. There are systems for other diseases and other types of images under development; however, this process depends on the availability of an accurate database. The author helped in the development of the databases for such systems and management of the clinical trial that resulted in the FDA-PMA of the system that detects findings consistent with lung cancer. The characteristics of the database used will be described. Further, a woman's risk of developing breast cancer differs from those of other women. Risk can be high, average, or low. There are now pharmaceuticals that decrease the risk that women, as a group, will develop breast cancer and it has been suggested that dietary changes could have similar effects. The pharmaceutical agents, though, have some associated side effects, and it is clinically important to determine whether these agents have decreased an individual woman's risk of breast cancer. In vivo imaging biomarkers of risk and successful risk reduction are therefore sought, but the information on possible in vivo imaging biomarkers is less mature than activities in CAD. Bioinformatics will be an important contributor to this in vivo imaging biomarker development.