Making Reliable Diagnoses with Machine Learning: A Case Study
In the past decades Machine Learning tools have been successfully used in several medical diagnostic problems. While they often significantly outperform expert physicians (in terms of diagnostic accuracy, sensitivity, and specificity), they are mostly not used in practice. One reason for this is that it is dificult to obtain an unbiased estimation of diagnose’s reliability. We propose a general framework for reliability estimation, based on transductive inference. We show that our reliability estimation is closely connected with a general notion of significance tests. We compare our approach with classical stepwise diagnostic process where reliability of diagnose is presented as its post-test probability. The presented approach is evaluated in practice in the problem of clinical diagnosis of coronary artery disease, where significant improvements over existing techniques are achieved.
Keywordsmachine learning medical diagnosis reliability estimation stepwise diagnostic process coronary artery disease
Unable to display preview. Download preview PDF.
- J. Dougherty, R. Kohavi, and M. Sahami. Supervised and unsupervised discretization of continuous features. In Proc. ICML’95, pages 194–202. Morgan Kaufmann, 1995.Google Scholar
- A. Gammerman, V. Vovk, and V. Vapnik. Learning by transduction. In Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence, pages 148–155, Madison, Wisconsin, 1998.Google Scholar
- A. L. Gibbs and F. E. Su. On choosing and bounding probability metrics. Technical report, Cornell School of Operations Research, 2000.Google Scholar
- M. Kukar. Estimating classifications’ reliability. PhD thesis, University of Ljubljana, Faculty of Computer and Information Science, Ljubljana, Slovenia, 2001. In Slovene.Google Scholar
- M. Kukar and C. Grošelj. Machine learning in stepwise diagnostic process. In Proc. Joint European Conference on Artificial Intelligence in Medicine and Medical Decision Making, pages 315–325, Aalborg, Denmark, 1999.Google Scholar
- M. Kukar, C. Grošelj, I. Kononenko, and J. Fettich. An application of machine learning in the diagnosis of ischaemic heart disease. In Proc. Sixth European Conference of AI in Medicine Europe AIME’97, pages 461–464, Grenoble, France, 1997.Google Scholar
- M. Olona-Cabases. The probability of a correct diagnosis. In J. Candell-Riera and D. Ortega-Alcalde, editors, Nuclear Cardiology in Everyday Practice, pages 348–357. Kluwer, 1994.Google Scholar
- B. H. Pollock. Computer-assisted interpretation of noninvasive tests for diagnosis of coronary artery disease. Cardiovasc. Rev. Rep. 4, pages 367–375, 1983.Google Scholar
- C. Saunders, A. Gammerman, and V. Vovk. Transduction with confidence and credibility. In Proceedings of the International Joint Conference on Artificial Intelligence, Stockholm, Sweden, 1999.Google Scholar
-  V. Vovk, A. Gammerman, and C. Saunders. Machine learning application of algorithmic randomness. In Proceedings of the 16th International Conference on Machine Learning (ICML’99), Bled, Slovenija, 1999.Google Scholar