Neural network and fuzzy clustering approach for automatic diagnosis of coronary artery disease in nuclear medicine

We investigated feasibility of using fuzzy clustering and artificial neural network to predict coronary artery disease (CAD) in acute phase from planar and gated SPECT nuclear medicine images. We developed an automatic computerized scheme that helps physicians diagnose coronary artery disease based on 99Tc-Sestamibi myocardial perfusion images. Our study consisted of two separate studies with respect to patient population and imaging method. The first study included 58 subjects (30 male, 28 female) studies using planar rest and stress imaging and a second patient subset of 115 subjects (61 male, 54 female) using gated rest/stress SPECT imaging. After the myocardial perfusion scans, patients also had coronary angiography within three months of the imaging. Signal-to-noise ratio was improved by segmentation of myocardium from its background in both studies using fuzzy clustering with the Picard iteration algorithm. We extracted a set of adaptive features consistent with nature of nuclear medicine imaging and myocardium anatomy. Features were optimized and selected based on maximum separation in multidimensional feature space. A back-propagation artificial neural network (ANN) classifier was trained and tested for each study using the optimal features and the results of coronary angiographies as input and outputs, respectively. ANNs were trained, optimized, and tested using leave-one-out and Poh's Implementation of Weigned-Rumelhart-Huberman (PIWRH) methods, to diagnose the normal and abnormal patients based on their coronary angiograms. The performances of the optimal ANNs were analyzed by receiver operator characteristic (ROC) method. Results of ANN in the first study were compared to those of the physicians in nuclear medicine ward and two other physicians using ROC method. Results of ANN for the second study were compared to those of the nuclear medicine ward using ROC method. Both subsets demonstrate that the proposed method outperforms visual diagnosis and is therefore a useful adjunct for CAD diagnosis from planar and gated SPECT images.

[1]  Ronald J. Jaszczak,et al.  ROC evaluation of SPECT myocardial lesion detectability with and without single iteration non-uniform Chang attenuation compensation using an anthropomorphic female phantom , 1998 .

[2]  Milton J. Guiberteau.  Nuclear cardiovascular imaging : current clinical practice , 1990 .

[3]  Hean Lee Poh,et al.  A neural network approach for marketing strategies research and decision support , 1991 .

[4]  H Fujita,et al.  Application of artificial neural network to computer-aided diagnosis of coronary artery disease in myocardial SPECT bull's-eye images. , 1992, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[5]  C E Floyd,et al.  Artificial neural network for diagnosis of acute pulmonary embolism: effect of case and observer selection. , 1995, Radiology.

[6]  Ronald J. Jaszczak,et al.  ROC evaluation of SPECT myocardial lesion detectability with and without nonuniform attenuation compensation using an anthropomorphic female phantom , 1997, 1997 IEEE Nuclear Science Symposium Conference Record.

[7]  David M. Skapura,et al.  Neural networks - algorithms, applications, and programming techniques , 1991, Computation and neural systems series.

[8]  C D Cooke,et al.  Diagnostic performance of an expert system for the interpretation of myocardial perfusion SPECT studies. , 2001, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[9]  L Edenbrandt,et al.  Automated Interpretation of Myocardial Spect Perfusion Images Using Artificial Neural Networks , 2022 .

[10]  J L Ritchie,et al.  Thallium-201 myocardial imaging: an interinstitutional study of observer variability. , 1978, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[11]  M E Hosseini-Ashrafi,et al.  Pre-optimization of radiotherapy treatment planning: an artificial neural network classification aided technique. , 1999, Physics in medicine and biology.

[12]  Daniel S. Marley,et al.  Clinical Decision Making , 1991 .

[13]  Miroslaw Pawlak,et al.  On Image Analysis by Moments , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  J M Boone,et al.  Neural networks in radiology: an introduction and evaluation in a signal detection task. , 1990, Medical physics.

[15]  C. V. Jawahar,et al.  Investigations on fuzzy thresholding based on fuzzy clustering , 1997, Pattern Recognit..

[16]  Gürsel Serpen,et al.  A novel algorithm for classification of SPECT images of a human heart , 1996, Proceedings Ninth IEEE Symposium on Computer-Based Medical Systems.

[17]  J. Sau,et al.  Automated detection of the left ventricular region in gated nuclear cardiac imaging , 1996, IEEE Transactions on Biomedical Engineering.

[18]  C. V. Jawahar,et al.  Incorporation of gray-level imprecision in representation and processing of digital images , 1996, Pattern Recognit. Lett..

[19]  K. Brown,et al.  The site of acute myocardial infarction is related to the coronary territory of transient defects on prior myocardial perfusion imaging , 1996, Journal of nuclear cardiology : official publication of the American Society of Nuclear Cardiology.

[20]  Evor L. Hines,et al.  Diagnosis of single photon emission tomography images using artificial neural networks , 1992 .

[21]  Carl G. Looney,et al.  Pattern recognition using neural networks: theory and algorithms for engineers and scientists , 1997 .

[22]  G V Heller,et al.  Relation between perfusion defects on stress technetium-99m sestamibi SPECT scintigraphy and the location of a subsequent acute myocardial infarction. , 1996, The American journal of cardiology.

[23]  K L Lam,et al.  Computerized classification of malignant and benign microcalcifications on mammograms: texture analysis using an artificial neural network. , 1997, Physics in medicine and biology.

[24]  J Soler-Soler,et al.  Relationship between the location of the most severe myocardial perfusion defects, the most severe coronary artery stenosis, and the site of subsequent myocardial infarction. , 2001, Journal of nuclear medicine : official publication, Society of Nuclear Medicine.

[25]  G.D. Tourassi,et al.  Application of neural networks to lesion detection in SPECT , 1991, Conference Record of the 1991 IEEE Nuclear Science Symposium and Medical Imaging Conference.

[26]  Keinosuke Fukunaga,et al.  Effects of Sample Size in Classifier Design , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[27]  W. Santamore,et al.  Can coronary angiography predict the site of a subsequent myocardial infarction in patients with mild-to-moderate coronary artery disease? , 1988, Circulation.

[28]  Hiroshi Fujita,et al.  Neural network approach for the computer-aided diagnosis of coronary artery diseases in nuclear medicine , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[29]  Domenico Cianflone,et al.  Efficient performance of neural network models as artificial intelligence prediction tools in cardiology , 1990, [1990] Proceedings Computers in Cardiology.

[30]  Georgia D. Tourassi,et al.  Artificial neural network for diagnosis of acute pulmonary embolism: Effect of case and observer selection , 1995 .

[31]  B. Dumitresco,et al.  Contouring blood pool myocardial gated SPECT images with a neural network leader segmentation and a decision-based fuzzy logic , 1997, Proceedings of 6th International Fuzzy Systems Conference.

[32]  Sarunas Raudys,et al.  On Dimensionality, Sample Size, Classification Error, and Complexity of Classification Algorithm in Pattern Recognition , 1980, IEEE Transactions on Pattern Analysis and Machine Intelligence.