Skip to main content
SpringerLink
Log in

A feed forward neural network for classification of bull's-eye myocardial perfusion images

  • Original Article
  • Published:
European Journal of Nuclear Medicine Aims and scope Submit manuscript

Abstract

Identification of hypoperfused areas in myocardial perfusion single-photon emission tomography studies can be aided by bull's-eye representation of raw counts, lesion extent and lesion severity, the latter two being produced by comparison of the raw bull's-eye data with a normal data base. An artificial intelligence technique which is presently becoming widely popular and which is particularly suitable for pattern recognition is that of artificial neural network. We have studied the ability of feed forward neural networks to extract patterns from bull's-eye data by assessing their capability to predict lesion presence without direct comparison with a normal data base. Studies were undertaken on both simulation data and on real stress-rest data obtained from 410 male patients undergoing routine thallium-201 myocardial perfusion scintigraphy. The ability of trained neural networks to predict lesion presence was quantified by calculating the areas under receiver operating characteristic curves. Figures as high as 0.96 for non-preclassified patient data were obtained, corresponding to an accuracy of 92%. The results demonstrate that neural networks can accurately classify patterns from bull's-eye myocardial perfusion images and detect the presence of hypoperfused areas without the need for comparison with a normal data base. Preliminary work suggests that this technique could be used to study perfusion patterns in the myocardium and their correlation with clinical parameters.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. Caldwell JH, Williams DL, Harp GD, Stratton JR, Richie JL. Quantitation of size of relative myocardial perfusion defect by single-photon emission computed tomography.Circulation 1984; 70: 1048–1056.

    Google Scholar 

  2. Garcia EV, van Train K, Maddahi J, Prigent F, Friedman J, Areeda J, Waxman A, Berman DS. Quantification of rotational thallium-201 myocardial tomography.J Nucl Med 1985; 26: 17–26.

    Google Scholar 

  3. DePasquale EE, Nody AC, DePuey EG, Garcia EV, Pilcher G, Bredlau C, Roubin G, Gober A, Gruentzig A, D'Amato P, Berger H. Quantitative rotational thallium-201 tomography for identifying and localizing coronary artery disease.Circulation 1988; 77: 316–327.

    Google Scholar 

  4. Wasserman PD.Neural computing. Theory and practice. New York: Van Nostrand Reinhold, 1989.

    Google Scholar 

  5. Eberhart RC, Dobbins RW.Neural network PC tools: a practical guide. San Diego: Academic Press, 1990.

    Google Scholar 

  6. Clark JW. Neural network modelling.Phys Med Biol 1991; 36: 1259–1317.

    Google Scholar 

  7. Miller AS, Blott BH, Hames TK. Review of neural network applications in medical imaging and signal processing.Med Biol Eng Comput 1992; 30: 449–1164.

    Google Scholar 

  8. Scott R. Artificial intelligence: its use in medical diagnosis.J Nucl Med 1993; 34: 510–514.

    Google Scholar 

  9. Boone JM, Gross GW, Greco-Hunt V. Neural networks in radiologic diagnosis. I. Introduction and illustration.Invest Radiol 1990; 25: 1012–1016.

    Google Scholar 

  10. Boone JM, Sigillito VG, Shaber GS. Neural networks in radiology: an introduction and evaluation in a signal detection task.Med Phys 1990; 17: 234–241.

    Google Scholar 

  11. Guerriere MRJ, Detsky AS. Neural networks: what are they? [editorial].Ann Intern Med 1991; 115: 906–907.

    Google Scholar 

  12. Ezquerra N, Garcia E. Artificial intelligence gives computer new role as imaging problem-solver.Diagn Imag 1985; 7: 195–200.

    Google Scholar 

  13. Makhoul J. Artificial neural networks.Invest Radiol 1990; 25: 748–750.

    Google Scholar 

  14. Shufflebarger CM, Young W. What is a neural network? [editorial].Ann Emerg Med 1992; 21: 1461–1462.

    Google Scholar 

  15. Allen J, Murray A. Development of a neural network screening aid for diagnosing lower limb peripheral vascular disease from photoelectric plethysmography pulse waveforms.Physiol Meas 1993; 14: 13–22.

    Google Scholar 

  16. Akay M. Noninvasive diagnosis of coronary artery disease using a neural network algorithm.Biol Cybern 1992; 67: 361–367.

    Google Scholar 

  17. Baxt WG. Use of an artificial neural network for data analysis in clinical decision making: the diagnosis of acute coronary occlusion.Neural Computation 1990; 2: 480–489.

    Google Scholar 

  18. Baxt WG. Use of an artificial neural network for the diagnosis of myocardial infarction.Ann Intern Med 1991; 115: 843–848.

    Google Scholar 

  19. Baxt WG. Analysis of the clinical variables driving decision in an artificial neural network trained to identify the presence of myocardial infarction.Ann Emerg Med 1992; 21: 1439–1444.

    Google Scholar 

  20. Bounds DG, Lloyd PJ, Mathew BG. A comparison of neural network and other pattern recognition approaches to the diagnosis of low back disorders.Neural Networks 1990; 3: 583–591.

    Google Scholar 

  21. Nelson TR, Boone JM. Visualization of myocardial activation sequences with neural network-based electrocardiographic localization: simulation results [abstract].RSNA 1992; 56.

  22. Asada N, Doi K, MacMahon H, Montner SM, Giger ML, Abe C, Wu Y. Potential usefulness of an artificial neural network for differential diagnosis of interstitial lung diseases: pilot study.Radiology 1990; 177: 857–860.

    Google Scholar 

  23. Fukuda H, Usuki N, Salwai S, Nakajima H, Miyamoto T, Inoue Y Usefulness of an artificial neural network for assessing the ventricular size [abstract].RSNA 1992; 56.

  24. Gross GW, Boone JM, Greco-Hunt V, Greenberg B. Neural networks in radiologic diagnosis. II. Interpretation of neonatal chest radiographs.Invest Radiol 1990; 25: 1017–1023.

    Google Scholar 

  25. Piraino DW, Amartur SC, Richmond BJ, Schils JP, Thome JM, Belhobek GH, Schlueter MD. Application of an artificial neural network in radiographic diagnosis.J Dig Imag 1991; 4: 226–232.

    Google Scholar 

  26. Scott JA, Palmer EL. Neural network analysis of ventilation-perfusion lung scans.Radiology 1993; 186: 661–664.

    Google Scholar 

  27. Wu Y, Doi K, Giger ML, Metz CE, Zhang W. Detection of lung nodules on digital radiographs: comparison of artificial neural networks and discriminant analysis [abstract].RSNA 1992; 56.

  28. Chan KK, Hayrapetian AS, Lau CC, Lufkin R. Neural network segmentation of double-echo MR images [abstract].RSNA 1992; 157.

  29. Kim JH, Min BG, Han MC, Lee CW. Computer-assisted detection of lung nodules by using artificial neural net [abstract].RSNA 1992; 56.

  30. Kippenham JS, Barker WW, Pascal S, Nagel J, Duara R. Evaluation of a neural-network classifier for PET scans of normal and Alzheimer's disease subjects.J Nucl Med 1992; 33: 1459–1467.

    Google Scholar 

  31. Floyd CE, Bowsher JE, Munley MT, Tourassi GD, Baydush AH, Coleman RE. Neural network for quantitative reconstruction of SPECT images [abstract].J Nucl Med 1991; 32: 936.

    Google Scholar 

  32. Garg S, Floyd CE. Neural network localization of pulmonary nodules on digital chest radiographs [abstract].RSNA 1992; 157.

  33. Floyd CE, Tourassi GD. An artificial neural network for lesion detection on single-photon emission computed tomographic images.Invest Radial 1992; 27: 667–672.

    Google Scholar 

  34. Tourassi GD, Floyd CE, Coleman RE. Detection and localization of cold lesions on SPECT images with artificial neural networks [abstract].RSNA 1992; 157.

  35. Fujita H, Katafuchi T, Uehara T, Nishimura T. Application of artificial neural network to computer-aided diagnosis of coronary artery disease in myocardial SPECT bull's eye images.J Nucl Med 1992; 33: 272–276.

    Google Scholar 

  36. Chan KH, Johnson KA, Becker JA, Satlin A, Mendelson J, Garada B, Holman BL. A neural network classifier for cerebral perfusion imaging.J Nucl Med 1994; 35: 771–774.

    Google Scholar 

  37. Datz FL, Rosenberg C, Gabor FV, Christian PE, Gullberg GT, Ahluwalia R, Morton KA. The use of computer-assisted diagnosis in cardiac perfusion nuclear medicine studies: a review (part 3).J Dig Imag 1993; 6: 67–80.

    Google Scholar 

  38. Wang DC, Juni JE. Artificial neural network interpretation of cardiac stress thallium studies [abstract].RSNA 1992; 283.

  39. Hamilton D, Riley PJ, Miola UJ, Amro A. Neural network analysis of T1-201 SPECT bullseye images [abstract]. Proc: 1st International Congress of Nuclear Cardiology, Cannes, France, April 1993.

  40. Riley P, Hamilton D, Miola UJ, Amro A. Neural network discrimination of myocardial ischaemia and infarction in201Tl SPECT imaging [abstract].Br J Radiol 1993; C75.

  41. Metz CE. Basic principles of ROC analysis.Semin Nucl Med 1978; 8: 283–298.

    Google Scholar 

  42. Metz CE. ROC methodology in radiologic imaging.Invest Radiol 1986; 21: 720–733.

    Google Scholar 

  43. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristics (ROC) curve.Radiology 1982; 143: 29–36.

    Google Scholar 

  44. Meistrell ML. Evaluation of neural network performance by receiver operating characteristic (ROC) analysis: examples from the biotechnology domain.Comput Methods Programs Biomed 1990; 32: 73–80.

    Google Scholar 

  45. Baxt WG. Improving the accuracy of an artificial neural network using multiple differently trained networks:Neural Computation 1992; 4: 772–780.

    Google Scholar 

  46. Mundler O, Grousset C, Birkui P, Pauchet M. Two years follow up of patients with “false positive” exercise or dipyridamole thallium 201 myocardial SPECT [abstract].Eur J Nucl Med 1989; 15: 416.

    Google Scholar 

  47. Rosen SD, Camici PG. Syndrome X: radionuclide studies of myocardial perfusion in patients with chest pain and normal coronary arteriograms.Eur J Nucl Med 1992; 19: 311–314.

    Google Scholar 

  48. Solot G, Hermans J, Merlo P, Chaudron J-M, Luwaert R, Cheron P, Bodart F, Beauduin M. Correlation of99Tcm-sestamibi SPECT with coronary angiography in general hospital practice.Nucl Med Commun 1993; 14: 23–29.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Medical Physics, Clinical and Bio-Engineering, Military Hospital, PO Box 7897, 11159, Riyadh, Kingdom of Saudi Arabia

    David Hamilton, Peter J. Riley & Ueber J. Miola

  2. Armed Forces Cardiac Centre, Riyadh, Kingdom of Saudi Arabia

    Ahmed A. Amro

Authors
  1. David Hamilton
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Peter J. Riley
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ueber J. Miola
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Ahmed A. Amro
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Hamilton, D., Riley, P.J., Miola, U.J. et al. A feed forward neural network for classification of bull's-eye myocardial perfusion images. Eur J Nucl Med 22, 108–115 (1995). https://doi.org/10.1007/BF00838939

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF00838939

Key words

Search

Navigation