A Hybrid Approach to Breast Cancer Diagnosis

  • M. Sordo
  • H. Buxton
  • D. Watson
Part of the International Series in Intelligent Technologies book series (ISIT, volume 16)

Abstract

In vivo 31P Magnetic Resonance Spectroscopy (MRS) is a non-invasive technique for the observation of phosphorus-containing metabolites and intracellular pH. MRS plays an important role in the investigation of cell biochemistry and offers a reliable means for detection of metabolic changes in breast tissue. However, the scarcity of 31P MRS data and the complexity of interpretation of relevant physiological information impose extra demands that preclude the applicability of most statistical and machine learning techniques developed so far. To overcome such constraints, we propose Knowledge-Based Artificial Neural Networks (KBANNs) [1], a hybrid methodology that combines knowledge from a domain in the form of simple rules with connectionist learning. This combination allows the use of small sets of data (typical of medical diagnosis tasks) to train the network. The initial structure is set from the dependencies of a set of known domain rules and it is only necessary to refine these rules by training. In this chapter, we present KBANNs with a topology derived from knowledge elicited from the domain of metabolic features of normal and malignant mammary tissues. KBANN performance is assessed over the classification of 26in vivo 31P spectra of normal and cancerous breast tissues. Results confirm the suitability of KBANNs as a computational aid capable of classifying complex and limited data in a medical domain. The present study is part of an ongoing investigation into normal and abnormal breast physiology, which may help in the non-invasive early detection of breast cancer [2], [3].

Keywords

Leukemia Adenosine Tamoxifen Phosphocreatine Phosphocholine 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Towell, G.G. (1991)Symbolic Knowledge and Neural Networks: Insertion Refinement and ExtractionPh.D. thesis, University of Wisconsin, Madison.Google Scholar
  2. 2.
    Sordo, M., Buxton, H., Watson, D., Collins, D., Ronen, S., Leach, M., and Payne, G. (1998), “KBANNs for classification of normal31P MRS based on hormonedependent changes during the menstrual cycle,”Fourth International Conference on Neural Networks and their ApplicationsMarseilles, France.Google Scholar
  3. 3.
    Sordo, M., Buxton, H., and Watson, D. (1999), “KBANNs and the classification of3 tP MRS of malignant mammary tissues,”Ninth International Conference on Artificial Neural NetworksEdinburgh, U.K.Google Scholar
  4. 4.
    Baxt, W.G. (1995), “Application of artificial neural networks to clinical medicine,”Lancetvol. 346, pp. 1135–1138.CrossRefGoogle Scholar
  5. 5.
    Sordo, M. (1999)A Neurosymbolic Approach to the Classification of Scarce and Complex DataD.Phil. thesis, University of Sussex, Brighton, U.K.Google Scholar
  6. 6.
    Leach, M. (1994), “Magnetic resonance spectroscopy applied to clinical oncology,”Technology and Health Carevol. 2, pp. 235–246.Google Scholar
  7. 7.
    Kuipers, B. and Kassier,J. (1987), “Konwledge acquisition by analysis of verbatim protocols, ” chapter in Knowledge Acquisitionfor Expert Systems. A Practical Handbookpp. 45–70, Plenum Press.Google Scholar
  8. 8.
    Fox, J., Myers, C., Greaves, M., and Pegram, S. (1987), “A systematic study of knowledge base refinement in the diagnosis of leukemia,” chapter inKnowledge Acquisition for Expert Systems. A Practical Handbookpp. 73–89, Plenum Press.Google Scholar
  9. 9.
    Johnson, L., and Johnson, N. (1987), “Knowledge elicitation involving teachback interviewing,” chapter inKnowledge Acquisition for Expert Systems. A Practical Handbookpp. 73–89, Plenum Press.Google Scholar
  10. Sordo, M., Buxton, H., and Watson, D. (1998), “A knowledge base for classification of normal breast31P MRS,” in Ifeachor, E., Sperduti, A., and Starita, A. (Eds.),3rd International Conference on Neural Networks and Expert Systems in Medicine and HealthcarePisa, Italy.Google Scholar
  11. 11.
    Mitchell, T.M., Keller, R.M., and KedarCabelli, S.T. (1986), “Explanationbased generalization: a unifying view,”Machine Learningvol. 1, pp. 47–80.Google Scholar
  12. 12.
    Rumelhart, D., McClelland, J., et al. (1986)Parallel Distributed Processing. Explorations in the Microstructure of Cognition Vol. 1: FoundationsM.I.T.Google Scholar
  13. 13.
    Towell, G., Shavlik, J., and Noordewier, M. (1990), “Refinement of approximate domain theories by knowledgebased neural networks,”VIII National Conference on Artificial Intelligencevol. 2, pp. 861–866.Google Scholar
  14. 14.
    Towell, G. and Lehrer, R. (1993), “A knowledgebased model of geometry learning,” in Hanson, S., Cowan, J., and Giles, C. (Eds.)Advances in Neural Information Processing Systemsvol. 5, pp. 887–894, Morgan Kauffmann, San Mateo, CA.Google Scholar
  15. 15.
    Towell, G., and Shavlik, J.W. (1989), “Combining Explanation- Based and Neural Learning: an Algorithm and Empirical Results,”Google Scholar
  16. 16.
    Towell, G. and Shavlik, J.W. (1994), “Knowledgebased a rtificial neural networks,”Artificial Intelligencevol. 70, nos. 1–2, pp. 119–165.Google Scholar
  17. 17.
    Towell, G.G. and Shavlik, J.W. (1992), “Using symbolic learning to improve knowledgebased neural networks,”X National Conference on Artificial Intelligencepp. 177–182.Google Scholar
  18. 18.
    Smith, T., Glaholm, J., Leach, M., Collins, D., et al. (1991), “A comparison of in vivo and in vitro31P NMR spectra from human breast tumours: variations in phospholipid metabolism,”British Journal of Cancervol. 63, no. 4, pp. 514–516.CrossRefGoogle Scholar
  19. 19.
    Bishop, R. and Bell, R. (1988), “Assembly of phospholipids into cellular membranes: biosynthesis, transmembrane movement and intracellular translocation,”Ann. Rev. Cell Biol.vol. 4, pp. 579–610.Google Scholar
  20. 20.
    Pelech, S. and Vance, D. (1989), “Signal transduction via phophatidylcholine cycles,”TIBSvol. 14, pp. 28–30.Google Scholar
  21. 21.
    RuizCabello, J. and Cohen, J.S. (1992), “Phospholipid meta - bolites as indicators of cancer cell function,”NMR in Biomedicinevol. 5, pp. 226–233.CrossRefGoogle Scholar
  22. 22.
    Kalra, R., Wade, K., Hands, L., et al. (1993), “Phosphomonoester is associated with proliferation in human breast cancer: a31P MRS study,”British Journal of Cancervol. 67, pp. 1145–1153.CrossRefGoogle Scholar
  23. 23.
    Su, B., Kappler, F., Szwergold, B., and Brown, T. (1993), “Identification of a putative tumor marker in breast and colon cancer,”Cancer Researchvol. 53, pp. 1751–1754.Google Scholar
  24. 24.
    Leach, M., Verrill, M., GlaholmJ.Smith, T., Collins, D., et al. (1998), “Measurements of human breast cancer using magnetic resonance spectroscopy: a review of clinical measurements and areport of localised 31P measurements of response to treatment,” NMR in Biomedicinevol. 11, no. 7, pp. 314–340.CrossRefGoogle Scholar
  25. 25.
    Ting, Y., Sherr, D., and Degani, H. (1996), “Variations in energy phospholipid metabolism in normal and cancer human mammary epithelial cells,”Anticancer Researchvol. 16, pp. 1381–1388.Google Scholar
  26. 26.
    Merchant, T., Meneses, P., Gierke, L., Otter, W.D., and Glonek, T. (1991), “31P magnetic resonance phospholipid profiles of neo-plastic human breast tissues,”British Journal of Cancervol. 63, pp. 693–698.CrossRefGoogle Scholar
  27. 27.
    Cohen, J., Lyon, R., Chen, C., Faustino, P., Batist, G., et al. (1986), “Differences in phosphate metabolite levels in drugsensitive and resistant human breast cancer cell lines determined by31P magnetic resonance spectroscopy,”Cancer Researchvol. 46, pp. 4087–4090.Google Scholar
  28. 28.
    Daly, P. and Cohen, J. (1989), “Magnetic resonance spectroscopy of tumors and potential in vivo clinical applications: a review,”Cancer Researchvol. 49, pp. 770–779.Google Scholar
  29. 29.
    Oberhaensli, R., Hilton-Jones, D., Bore, P., et al. (1986), “Bio - chemical investigation of human tumours in vivo with phosphorus31 magnetic resonance spectroscopy,”Lancetvol. 2, pp. 8–11.CrossRefGoogle Scholar
  30. 30.
    Steen, R. (1989), “Response of solid tumors to chemotherapy monitored by in vivo31P nuclear magnetic resonance spectroscopy: a review,”Cancer Researchvol. 49, pp. 4075–4085.Google Scholar
  31. 31.
    Negendank, W. (1992), “Studies of human tumors by MRS: a review,”NMR in Biomedicinevol. 5, pp. 303–324.CrossRefGoogle Scholar
  32. 32.
    Gadian, D. (1995)NMR and Its Applications to Living Systems(2nd edition), Oxford University Press.Google Scholar
  33. 33.
    Kidd, A.L. (1987), “Knowledge acquisition — an introductory framework,”chapter inKnowledge Acquisition for Expert Systems. A Practical Handbookpp. 1–16, Plenum Press.Google Scholar
  34. 34.
    Redmond, O., Bell, E., Stack, J., Dervan, P., Carney, D., Hurson, B., and Ennis, J. (1992), “Tissue characterization and assessment of preoperative chemotherapeutic response in musculoskeletal tumors by in vivo31P magnetic resonance spectroscopy,”Magnetic Resonance in Medicinevol. 27, pp. 226–237.CrossRefGoogle Scholar
  35. 35.
    Degani, H., Horowitz, A., and Itzchak, Y. (1986), “Breast tumors: evaluation with P31 MR spectroscopy,”Radiologyvol. 161, pp. 53–55.Google Scholar
  36. 36.
    Glaholm, J., Leach, M., Collins, D., Mansi, J., et al. (1989), “Invivo31P magnetic resonance spectroscopy for monitoring treatment response in breast cancer,”Lancetvol. 8650, no. 1, pp. 1326–1327.CrossRefGoogle Scholar
  37. 37.
    Twelves, C., Porter, D., Lowry, M., Dobbs, N., et al. (1994), “Phosphorus31 metabolism of postmenopausal breast cancer studied in vivo by magnetic resonance spectroscopy,”British Journal of Cancervol. 69, pp. 1151–1156.CrossRefGoogle Scholar
  38. 38.
    Ng, T., Evanochko, W., Hiramoto, R., Ghanta, V., et al. (1982), “31P NMR spectroscopy of in vivo tumors,”Journal of Magnetic Resonancevol. 49, pp. 271–286.Google Scholar
  39. 39.
    Ross, B., Marshall, V., Smith, M., Bartlett, S., et al. (1984), “Monitoring response to chemotherapy of intact human tumours by 31P nuclear magnetic resonance,”Lancetno. 1, pp. 641–646.CrossRefGoogle Scholar
  40. 40.
    Payne, G., Dowsett, M., and Leach, M. (1994), “Hormone-dependent metabolic changes in the normal breast monitored noninvasively by31P magnetic resonance (MR) spectroscopy,”The Breastvol. 3, pp. 20–23.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2001

Authors and Affiliations

  • M. Sordo
  • H. Buxton
  • D. Watson

There are no affiliations available

Personalised recommendations