Advertisement

Detecting very early stages of dementia from normal aging with Machine Learning methods

  • William Rodman Shankle
  • Subramani Mani
  • Michael J. Pazzani
  • Padhraic Smyth
Knowledge Acquisition and Learning
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1211)

Abstract

We used Machine Learning (ML) methods to learn the best decision rules to distinguish normal brain aging from the earliest stages of dementia using subsamples of 198 normal and 244 cognitively impaired or very mildly demented (Clinical Dementia Rating Scale=0.5) persons. Subjects were represented by their age, education and gender, plus their responses on the Functional Activities Questionnaire (FAQ), the Mini-Mental Status Exam (MMSE), and the Ishihara Color Plate (ICP) tasks. The ML algorithms applied to these data contained within the electronic patient records of a medical relational database, learned rule sets that were as good as or better than any rules derived from either the literature or from domain specific knowledge provided by expert clinicians. All ML algorithms for all runs found that a single question from the FAQ, the forgetting rule, (“Do you require assistance remembering appointments, family occasions, holidays, or taking medications?”) was the only attribute included in all rule sets. CART's tree simplification procedure always found that just the forgetting rule gave the best pruned decision tree rule set with classification accuracy (93% sensitivity and 80% specificity) as high as or better than any other decision tree rule-set. Comparison with published classification accuracies for the FAQ and MMSE revealed that including some of the additional attributes in these tests actually worsen classification accuracy. Stepwise logistic regression using the FAQ attributes to classify dementia status confirmed that the forgetting rule gave a much larger odds ratio than any other attribute and was the only attribute included in all of the stepwise logistic regressions performed on 33 random samples of the data. Stepwise logistic regression using the MMSE attributes identified two attributes which occurred in all 33 runs and had by far the highest odds ratio. In summary, ML methods have discovered that the simplest and most sensitive screening test for the earliest clinical stages of dementia consists of a single question, the forgetting rule.

Keywords

Stepwise Logistic Regression Naive Bayesian Classifier Clinical Dementia Rate Scale Demented Subject Dementia Status 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    American Psychiatric Association, Washington, D. C. Diagnostic and Statistical Manual of Mental Disorders, 4 edition, 1994.Google Scholar
  2. 2.
    Ohmann C, Yang Q, Moustakis V, Lang K, and PJ van Elk. Machine learning techniques applied to the diagnosis of acute abdominal pain. In Pedro Barahona and Mario Stefanelli, editors, Lecture Notes in Artificial Intelligence: Artificial Intelligence in Medicine AIME95, volume 934, pages 276–281. Springer, 1995.Google Scholar
  3. 3.
    Cestnik G, Konenenko I, and Bratko I. Assistant-86: A knowledge-elicitation tool for sophisticated users. In Bratko I and Lavrac N, editors, Progress in Machine Learning, pages 31–45. Sigma Press, 1987.Google Scholar
  4. 4.
    Crum R.M, Anthony J.C, Bassett S.S, and Folstein M.F., Population-based norms for the mini-mental state examination by age and educational level. JAMA, 269(18):2386–2390, May 1993.Google Scholar
  5. 5.
    Heckerman D.E, Horvitz E.J, and Nathwani B.N., Towards normative expert systems: Part i the pathfinder project. Methods of Information in Medicine, (31):90–105, 1992.Google Scholar
  6. 6.
    Duda R.O and Hart P.E. Pattern Classification and Scene Analysis. John Wiley, New York, 1973.Google Scholar
  7. 7.
    Ernst R.L and Hay J.W., The u.s. economic and social costs of alzheimer's disease revisited. American Journal of Public Health, 84(8):1261–4, Aug 1994.Google Scholar
  8. 8.
    Fillenbaum G.G, Heyman A, Wilkinson W.E, and Haynes C.S., Comparison of two screening tests in alzheimer's disease—the correlation and reliability of the mini-mental state examination and the modified blessed test. Archives of Neurology, 44(9):924–7, Sep 1987.Google Scholar
  9. 9.
    Folstein M.F, Folstein S.E, and McHugh P.R., Mini-mental state-a practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3):189–98, Nov 1975.Google Scholar
  10. 10.
    Hoffman R.S., Diagnostic errors in the evaluation of behavioral disorders. JAMA, 248:225–8, 1982.Google Scholar
  11. 11.
    Shinobu Ishihara. Ishihara Tests for Colour-Blindness. Kanehara Shuppan, Ltd., Tokyo Japan, 1994.Google Scholar
  12. 12.
    Wyatt J. Lessons learned from the field trials of acorn, a chest pain advisor. In Barber B, Cao D, Qin D, and Wagner F, editors, Proceedings MedInfo, pages 111–115. Elsevier Scientific, 1989.Google Scholar
  13. 13.
    Brieman L., Friedman J.H., Olshen R.A., and Stone C.J. Classification and Regression Trees. Wadsworth, Belmont, 1984.Google Scholar
  14. 14.
    Gierl L. and Stengel-Rutkowski S., Integrating consultation and semi-automatic knowledge acquisition in a prototype-based architecture: Experiences with dysmorphic syndromes. Artificial Intelligence in Medicine, 6:29–49, 1994.Google Scholar
  15. 15.
    Nada Lavrac and Igor Mozetic. Second generation knowledge acquisition methods and their application to medicine. In Keravnou E, editor, Deep Models for Medical Knowledge Engineering, pages 177–198. Elsevier, New York, 1992.Google Scholar
  16. 16.
    Lubeck D.P, Mazonson T and Bowe P.D., Potential effect of tacrine on expenditures for alzheimer's disease. Medical Interface, 7(10):130–8, Oct 1994.Google Scholar
  17. 17.
    McCleary R, Shankle W.R, Mulnard R.A, and Dick M.B. Ishihara test performance and dementia. Journal of the Neurological Sciences, in press 1996.Google Scholar
  18. 18.
    R.S. Michalski, Mozetic I, Hong J, and Lavrac N. The multi-purpose incremental learning system aq15 and its testing application to three medical domains. In In Proceedings of the Fifth National Conference on Artificial Intelligence, pages 1041–1045, Philadelphia, PA, 1986. Morgan Kaufmann.Google Scholar
  19. 19.
    Morris J.C., The clinical dementia rating (cdr): current version and scoring rules. Neurology, 43(11):2412–4, Nov 1993.Google Scholar
  20. 20.
    Igor Mozetic and Bernhard Pfahringer. Improving diagnostic efficiency in kardio: Abstractions, constraint propagation and model compilation. In Keravnou E, editor, Deep Models for Medical Knowledge Engineering, pages 1–25. Elsevier, New York, 1992.Google Scholar
  21. 21.
    O'Connor D.W, Fertig A, Grande M.J, Hyde J.B, Perry J.R, Roland M.O, Silverman J.D and Wright S.K. Dementia in general practice: the practical consequences of a more positive approach to diagnosis. Br J Gen Pract, 43:185–8, 1993.Google Scholar
  22. 22.
    Oconnor D.W, Pollitt PA, Treasure F.P, Brook C.P.B, and Reiss B.B. The influence of education, social class and sex on mini-mental state scores. Psychological Medicine, 19:771–776, 1989.Google Scholar
  23. 23.
    Michael Pazzani and Dennis Kibler. The utility of knowledge in inductive learning. Machine Learning, (9):57–94, 1992.Google Scholar
  24. 24.
    Pfeffer R.I, Kurosaki T.T, Harrah C.H, Chance J.M, and Filos S. Measurement of functional activities in older adults in the community. J Gerontology, 37:323–9, 1982.Google Scholar
  25. 25.
    Quinlan J.R. C4.5: Programs for Machine Learning. Morgan Kaufmann, Los Altos, California, 1993.Google Scholar
  26. 26.
    Kohavi R, George John, Richard Long, David Manley, and Karl Pfleger. Mlc++: A machine learning library in c++. In Tools with Artificial Intelligence, pages 740–743. IEEE Computer Society Press, 1994.Google Scholar
  27. 27.
    Shankle W.R, Datta P, Dillencourt M, and Pazzani M. Improving dementia screening tests with machine learning methods. Alzheimer's Research, 2(3), Jun 1996.Google Scholar
  28. 28.
    Shortliffe E. Computer-Based Medical Consultations: MYCIN. Elsevier/North Holland, New York, 1976.Google Scholar
  29. 29.
    Welsh K.A, Butters N, Mohs R.C, Beekly D, Edland S, and Fillenbaum G. The consortium to establish a registry for alzheimer's disease (cerad. part v. a normative study of the neuropsychological battery. Neurology, 44(4):609–14, Apr 1994.Google Scholar
  30. 30.
    Williams T.F and Costa P.T. Recognition and initial assessment of alzheimer's disease and related dementias: Clinical practice guidelines. Technical report, Department of Health and Human Services, 1995.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • William Rodman Shankle
    • 1
  • Subramani Mani
    • 2
  • Michael J. Pazzani
    • 2
  • Padhraic Smyth
    • 2
  1. 1.Departments of Neurology and Information and Computer ScienceUniversity of California at IrvineIrvine
  2. 2.Department of Information and Computer ScienceUniversity of California at IrvineIrvine

Personalised recommendations