Purpose: To compare the diagnostic performance of an artificial neural network (ANN) with that of physicians m patients w-lth suspected pulmonary embolism (PE).Matcti& and M&&X An ANN was developed to predict PE by using findings from ventilation-perfusion lung scans and chest radiographs.First.the network was evaluated on 1 .Oh4 cases from the Prospectlvc Investigation of Pulmonary Embohsm Dlagnosls (PlOPED study that had a definitive angiographlc outcome.An uppet and lower bound of its diagnostic performance was provided depending on case difficulty Then. the network was tested on 104 patients with suspected PE in whom pulmonary anglography was essential for diagnosis The diagnostic performance L~I the ANN was compared with that of (a) two nuclear medicint physicians who read the scans for the needs of this study and (b) the nuclear medicine physicians who ongmally read lhc, scans.The effects of case and obsener selection on perfotmanic, were addressed.Results: The ANN outperformed the physlclanh when they used the PIOPED criteria lor categoric assessment, and it performed as well as the two study physicians on the basis of their probability assessments.Conclusion.The ANN call detect or exclude PE m a highly selected group of difficult cast\ ulth a consistency equivalent to that of ivery experienced $1) sicians.
No takes yet. Share an insight, caveat, or question.
Bindman et al. (1995) studied this question.