Three-dimensional mapping of acute ischemic regions using artificial neural networks and tagged MRI

Many methods for mapping ischemic myocardial regions by functional analysis have been suggested. However, the complicated relationship between myocardial function and perfusion, and the inherent limitations of the imaging techniques used, have led to a generally low mapping accuracy. The authors show herein, that highly accurate mapping can be obtained by combining tagged magnetic resonance imaging (MRI), three-dimensional (3-D) analysis, and artificial neural networks. Nine canine hearts with acute ischemia were studied using multiplanar tagged MRI. Twenty-four myocardial cuboids were tagged in each heart and reconstructed in 3-D at end diastole (ED) and end systole (ES). The cuboids were arranged in three slices approximately 1 cm thick and covered most of the left ventricle (LV). Transmural thickening and endocardial area strain were calculated for each cuboid. Applying a post-mortem (PM) analysis, the percent ischemia in each cuboid was estimated using monastral blue dye; the PM analysis served as a "gold standard". An artificial neural network (ANN), designed to estimate the percent ischemia in each cuboid from the functional indexes, was then created. The ANN "learned" the function-ischemia relationship in 192 cuboids taken from eight of the hearts and was asked to estimate the percent ischemia in the 24 cuboids of the ninth heart. The process was repeated nine times, each time using a different heart as test case. The average accuracy of mapping, i.e., the accuracy with which the ANN has mapped the normal and ischemic cuboids using the functional parameters, was 87.5%/spl plusmn/7.8 (s.d.). This accuracy was superior to the accuracy obtained by optimal thresholding of the same thickening (80.1%) and endocardial strain (76.9%) data.

[1]  G M Hutchins,et al.  Two‐dimensional Echocardiography and Infarct Size: Relationship of Regional Wall Motion and Thickening to the Extent of Myocardial Infarction in the Dog , 1981, Circulation.

[2]  N B Ingels,et al.  Measurement of Midwall Myocardial Dynamics in Intact Man by Radiography of Surgically Implanted Markers , 1975, Circulation.

[3]  J. Covell,et al.  Ventricular wall motion , 1987 .

[4]  J. Weiss,et al.  Impaired thickening of nonischemic myocardium during acute regional ischemia in the dog. , 1985, Circulation.

[5]  A. Buda,et al.  Nonischemic dysfunction at the lateral margins of ischemic myocardium , 1987 .

[6]  W. O'Dell,et al.  Calculation of three‐dimensional left ventricular strains from biplanar tagged MR images , 1992, Journal of magnetic resonance imaging : JMRI.

[7]  W. Brownell,et al.  Outer hair cell length changes in an external electric field. I. The role of intracellular electro-osmotically generated pressure gradients. , 1995, The Journal of the Acoustical Society of America.

[8]  E. Zerhouni,et al.  Noninvasive quantification of principal strains in normal canine hearts using tagged MRI images in 3-D. , 1993, The American journal of physiology.

[9]  E. Zerhouni,et al.  Human heart: tagging with MR imaging--a method for noninvasive assessment of myocardial motion. , 1988, Radiology.

[10]  Fred L. Bookstein,et al.  From Medical Images to the Biometrics of Form , 1986 .

[11]  L. Becker,et al.  Salvage of ischemic myocardium by prostacyclin during experimental myocardial infarction. , 1983, Journal of the American College of Cardiology.

[12]  R R Edelman,et al.  MRI evaluation of myocardial perfusion without a contrast agent using magnetization transfer , 1993, Magnetic resonance in medicine.

[13]  S Sideman,et al.  Discrimination between healthy and diseased hearts by spectra decomposition of their left ventricular three-dimensional geometry. , 1991, IEEE transactions on medical imaging.

[14]  Robert Plonsey,et al.  Stimulation of Spheroidal Cells - The Role of Cell Shape , 1976, IEEE Transactions on Biomedical Engineering.

[15]  Y. Fung,et al.  Transmural Myocardial Deformation in the Canine Left Ventricle: Normal in Vivo Three‐Dimensional Finite Strains , 1985, Circulation research.

[16]  S Sideman,et al.  Three-dimensional mapping of acute ischemic regions using MRI: wall thickening versus motion analysis. , 1990, The American journal of physiology.

[17]  L. Axel,et al.  MR imaging of motion with spatial modulation of magnetization. , 1989, Radiology.

[18]  H Azhari,et al.  A noninvasive comparative study of myocardial strains in ischemic canine hearts using tagged MRI in 3-D. , 1995, The American journal of physiology.

[19]  D Revel,et al.  Superparamagnetic iron oxide particles and positive enhancement for myocardial perfusion studies assessed by subsecond T1-weighted MRI. , 1993, Magnetic resonance imaging.

[20]  J F Ashmore,et al.  Ionic currents of outer hair cells isolated from the guinea‐pig cochlea. , 1992, The Journal of physiology.

[21]  S A MacKay,et al.  Methods for evaluating cardiac wall motion in three dimensions using bifurcation points of the coronary arterial tree. , 1983, Investigative radiology.

[22]  J. Weiss,et al.  Impaired Function of Salvaged Myocardium: Iwo-dimensional Echocardiographic Quantification of Regional Wall Thickening in the Open-chest Dog , 1983, Circulation.

[23]  W L Maughan,et al.  Abnormalities of Dynamic Ventricular Shape Change in Patients With Aortic and Mitral Valvular Regurgitation: Assessment by Fourier Shape Analysis and Global Geometric Indexes , 1988, Circulation research.

[24]  S. Vatner Correlation between Acute Reductions in Myocardial Blood Flow and Function in Conscious Dogs , 1980, Circulation research.

[25]  R Beyar,et al.  Quantification and validation of left ventricular wall thickening by a three-dimensional volume element magnetic resonance imaging approach. , 1990, Circulation.