Computer intuition: guiding scientific research in imaging and oral implantology

As the costs associated with clinical research and new drug development increase, it is incumbent upon us to develop alternative research methodologies. A new Computer Intuition (CI) program identifies literature-based evidence with the potential to generate a hypothesis that is most likely to be clinically confirmed regarding the cause of a disease or the questions being posed. The objective of this study was to demonstrate whether computer intuition can be used to guide scientific research in solving important biomedical questions. The establishment of CI as an alternative research methodology has the potential to accelerate the translation of basic science findings into clinical practice. The specific aim of this study was to develop a CI-driven hypothesis related to the controversial issue debating which radiographic imaging technology is most suitable for diagnostic purposes prior to the placement of oral implants. This hypothesis was then compared, in retrospect, with a known opinion established following a recent literature review on this issue. In his review of the literature, Frederiksen (1995) suggests that multiplanar reformatted computed tomography (CT) is the diagnostic imaging modality of choice prior to implant placement. To compare the CI hypothesis with Frederiksen's (1995) opinion, we subjected 34 relevant papers to CI analysis. The output consisted of clusters of important statements, phrases, and thought processes from the given dataset rated with the greatest potential to generate testable hypotheses. Both CI's output and Frederiksen (1995) indicated that CT scanning is the diagnostic imaging modality of choice prior to implant placement. Although the ultimate utility of CI is dependent on the successful testing of its derived hypothesis, this preliminary retrospective study suggests that CI might be useful in guiding scientific research.