Original ArticleAutomatic Fuzzy Classification of the Washout Curves From Magnetic Resonance First-Pass Perfusion Imaging After Myocardial InfarctionComte, Alexandre Msc*; Lalande, Alain PhD*; Cochet, Alexandre MD*; Walker, Paul M. PhD*; Wolf, Jean-Eric MD†; Cottin, Yves MD†; Brunotte, François MD* Author Information From the *Laboratoire de Biophysique, Faculté de Médecine, Université de Bourgogne, Dijon, France; and †Service de Cardiologie II, Centre Hospitalier Universitaire de Dijon, 2 Bd Maréchal de Lattre de Tassigny, 21034 Dijon Cedex, France. Received January 10, 2005 and accepted for publication, after revision, April 14, 2005. Reprints: Alexandre Comte, MSc, Laboratoire de Biophysique, Faculté de Médecine, Université de Bourgogne, BP 87900, 21079 Dijon Cedex, France. E-mail: [email protected]. Investigative Radiology: August 2005 - Volume 40 - Issue 8 - p 545-555 doi: 10.1097/01.rli.0000170448.31487.1b Buy Metrics Abstract Objectives: We sought to investigate the diagnostic ability of cardiac magnetic resonance imaging (MRI) perfusion in acute reperfused myocardial infarction. The study used fuzzy logic to automatically classify signal intensity-time curves from myocardial segments into 3 categories: normal, hypointense, and Hyperintense. Materials and Methods: Thirty-eight patients with myocardial infarction underwent short-axis cine-MRI and contrast-enhanced MRI to provide data on wall thickening and the transmural extent of infarction. Of these, 17 had a second cardiac MRI to ascertain the functional recovery in each segment. Results: The fuzzy logic based classification performs well (kappa= 0.87, P < 0.01) in comparison with a visual approach. Segments providing “hypo” curves do not recover (Δ = 0.11 SD = 0.96) whereas segments demonstrating the other curve types recover (Δ = 1 SD = 1.98 for “hyper” curves and Δ = 1.54 SD = 1.77 for “normal” curves). Conclusions: The proposed automatic signal intensity-time curve classification has a prognostic value when studying the functional recovery of pathologic segments and clearly identifies the no-reflow phenomenon known to induce poor recovery. © 2005 Lippincott Williams & Wilkins, Inc.