Classification of sporadic Creutzfeldt-Jakob disease based on clinical and neuropathological characteristics
- 122 Downloads
Creutzfeldt-Jakob disease (CJD) is a rare and fatal neurodegenerative disease of unknown cause. Patients are usually aged between 50 and 75 and typical clinical features include rapidly progressive dementia associated with myoclonus and a characteristic electroencephalographic pattern. Neuropathological examination reveals cortical spongiform change, hence the term ‘spongiform encephalopathy’. Several statistical techniques were applied to classify patients with sporadic CJD (sCJD), based on clinical and neuropathological investigation. We focus on the classification of neuropathologically confirmed sCJD patients. In order to obtain a classification rule that correctly classifies this type of patients and at the same time controls the overall error rate, we apply several classification techniques, which in general, produce comparable results. The boosting method produces the best results and the variable 14-3-3 protein in cerebrospinal fluid plays the most important role in the prediction of neuropathologically confirmed sCJD.
KeywordsBoosting methods Classification tree Sporadic Creutzfeldt-Jakob disease
- 17.Hastie T, Tibshirani R, Friedman J. The elements of statistical learning. Data mining, inference and prediction. New York: Springer-Verlag; 2001. p. 299–347.Google Scholar
- 18.Breiman L, Friedman JH, Olshen RA, Stone CJ. Classification and regression trees. New York: Chapman and Hall; 1984.Google Scholar
- 19.Freund Y, Schapire R. Experiments with a new boosting algorithm, Machine Learning: Proceedings of the Thirteenth International Conference, 1996. p. 148–56.Google Scholar
- 28.Therneau TM, Atkinson EJ. An introduction to recursive partitioning using the rpart routines. Technical Report 61, Department of Health Science Research, Mayo Clinic, Rochester; 1997.Google Scholar