Role of ventilation scintigraphy in diagnosis of acute pulmonary embolism: an evaluation using artificial neural networks

  • Eva Evander
  • Holger Holst
  • Andreas Järund
  • Mattias Ohlsson
  • Per Wollmer
  • Karl Åström
  • Lars Edenbrandt
Original Article


The purpose of this study was to assess the value of the ventilation study in the diagnosis of acute pulmonary embolism using a new automated method. Either perfusion scintigrams alone or two different combinations of ventilation/perfusion scintigrams were used as the only source of information regarding pulmonary embolism. A completely automated method based on computerised image processing and artificial neural networks was used for the interpretation. Three artificial neural networks were trained for the diagnosis of pulmonary embolism. Each network was trained with 18 automatically obtained features. Three different sets of features originating from three sets of scintigrams were used. One network was trained using features obtained from each set of perfusion scintigrams, including six projections. The second network was trained using features from each set of (joint) ventilation and perfusion studies in six projections. A third network was trained using features from the perfusion study in six projections combined with a single ventilation image from the posterior view. A total of 1,087 scintigrams from patients with suspected pulmonary embolism were used for network training. The test group consisted of 102 patients who had undergone both scintigraphy and pulmonary angiography. Performances in the test group were measured as area under the receiver operation characteristic curve. The performance of the neural network in interpreting perfusion scintigrams alone was 0.79 (95% confidence limits 0.71–0.86). When one ventilation image (posterior view) was added to the perfusion study, the performance was 0.84 (0.77–0.90). This increase was statistically significant (P=0.022). The performance increased to 0.87 (0.81–0.93) when all perfusion and ventilation images were used, and the increase in performance from 0.79 to 0.87 was also statistically significant (P=0.016). The automated method presented here for the interpretation of lung scintigrams shows a significant increase in performance when one or all ventilation images are added to the six perfusion images. Thus, the ventilation study has a significant role in the diagnosis of acute lung embolism.


Image processing Artificial neural networks Pulmonary embolism 



This study was supported by grants from the Swedish Medical Research Council (09893 and 10841).


  1. 1.
    Alderson PO, Rujanavech, N, Secker-Walker RH, McKnight RC. The role of133Xe ventilation studies in the scintigraphic detection of pulmonary embolism. Radiology 1976; 120:633–640.PubMedGoogle Scholar
  2. 2.
    Goris ML, Daspit SG. Krypton-81m ventilation scintigraphy for the diagnosis of pulmonary embolism. Clin Nucl Med 1981; 6:207–212.PubMedGoogle Scholar
  3. 3.
    Miniati M, Pistolesi M, Marini C, et al. Value of perfusion lung scan in the diagnosis of pulmonary embolism: results of the prospective investigative study of acute pulmonary embolism diagnosis (PISA-PED). Am J Respir Crit Care Med 1996; 154:1387–1393.PubMedGoogle Scholar
  4. 4.
    Holst H, Åström K, Järund A, et al. Automated interpretation of ventilation-perfusion lung scintigrams for the diagnosis of pulmonary embolism using artificial neural networks. Eur J Nucl Med 2000; 27:400–406.Google Scholar
  5. 5.
    Holst H, Måre K, Järund A, et al. An independent evaluation of a new method for automated interpretation of lung scintigrams using artificial neural networks. Eur J Nucl Med 2001; 28:33–38.Google Scholar
  6. 6.
    Tägil K, Evander E, Wollmer P, Palmer J, Jonson B. Efficient lung scintigraphy. Clin Physiol 2000; 20:95–100.CrossRefPubMedGoogle Scholar
  7. 7.
    Mandell CH. Scintillation camera lung imaging: an anatomic atlas and guide. New York: Grune and Stratton, 1976.Google Scholar
  8. 8.
    Cross SS, Harrison RF, Kennedy RL. Introduction to neural networks. Lancet 1995; 346:1075–1079.PubMedGoogle Scholar
  9. 9.
    Bishop CM. Neural networks for pattern recognition. Oxford: Oxford University Press, 1995.Google Scholar
  10. 10.
    Rumelhart DE, Hinton GE, Williams RJ. Learning internal representation by error propagation. In: Rumelhart DE, McCleland JL, eds. Parallel distributed processing: explorations in the microstructure of cognition. Cambridge, Mass: MIT Press/Bradford Books; 1986:318–363.Google Scholar
  11. 11.
    Hanson SJ, Pratt LY. Comparing biases for minimal network construction with backpropagation. In: Touretzky DS, ed. Advances in neural information processing systems. San Mateo, Ca.: Morgan Kaufmann; 1989:177–185.Google Scholar
  12. 12.
    Woods K, Bowyer KW. Generating ROC curves for artificial neural networks. IEEE Trans Med Imaging 1997; 16:329–337.CrossRefPubMedGoogle Scholar
  13. 13.
    Efron B. Computers and the theory of statistics: thinking the unthinkable. Siam Rev 1979; 21:460–480.Google Scholar
  14. 14.
    Wehrens R, Putter H, Buydens LMC. The bootstrap: a tutorial. Chemom Intell Lab Syst 2000; 54:35–52.CrossRefGoogle Scholar
  15. 15.
    Peterson C, Rögnvaldsson T, Lönnblad L. JETNET 3.0: a versatile artificial neural network package. Comp Phys Comm 1994; 81:185–220.CrossRefGoogle Scholar
  16. 16.
    McLaughlin GL, Burt RW, DePalma B, Gubler R. Effect of ventilation images on observer interpretation of lung perfusion examinations. Am J Roentgenol 1977; 128:1037–1038.Google Scholar
  17. 17.
    The PIOPED Investigators. Value of the ventilation/perfusion scan in acute pulmonary embolism: results of the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED). JAMA 1990; 263:2753–2759.PubMedGoogle Scholar
  18. 18.
    Stein PD, Terrin ML, Gottschalk A, Alavi A, Henry J. Value of ventilation/perfusion scans versus perfusion scans alone in acute pulmonary embolism. Am J Cardiol 1992; 69:1239–1241.PubMedGoogle Scholar
  19. 19.
    Tourassi GD, Frederick ED, Vittitoe NF, Coleman RE. Fractal texture analysis of perfusion lung scans. Comput Biomed Res 2000; 33:161–171.CrossRefPubMedGoogle Scholar
  20. 20.
    Tourassi GD, Frederick ED, Floyd CE Jr, Coleman RE. Multifractal texture analysis of perfusion lung scans as a potential diagnostic tool for acute pulmonary embolism. Comput Biol Med 2001; 31:15–25.CrossRefPubMedGoogle Scholar
  21. 21.
    Scott JA. Using artificial neural network analysis of global ventilation-perfusion scan morphometry as a diagnostic tool. Am J Roentgenol 1999; 173:943–948.Google Scholar
  22. 22.
    Scott JA, Palmer EL, Fischman AJ. How well can radiologists using neural network software diagnose pulmonary embolism? Am J Roentgenol 2000; 175:399–405.Google Scholar

Copyright information

© Springer-Verlag 2003

Authors and Affiliations

  • Eva Evander
    • 1
  • Holger Holst
    • 2
  • Andreas Järund
    • 2
  • Mattias Ohlsson
    • 3
  • Per Wollmer
    • 2
  • Karl Åström
    • 4
  • Lars Edenbrandt
    • 2
  1. 1.Department of Clinical PhysiologyUniversity Hospital, Lund UniversityLundSweden
  2. 2.Department of Clinical PhysiologyMalmö University HospitalMalmöSweden
  3. 3.Department of Theoretical PhysicsLund UniversityLundSweden
  4. 4.Department of MathematicsLund Institute of TechnologyLundSweden

Personalised recommendations