Author/Editor		Grabec, I; Ferkolj, I; Grošelj, D
Title		Prediction of clinical response to treatment in Crohn's disease
Type		članek
Source		In: Fonseca JM, editor. CIMED2005. Proceedings of the 2nd international conference on computational intelligence in medicine and healthcare; 2005 Jun 29-Jul 1; Lisbon. Lisbon: IEE healthcare technologies,
Publication year		2005
Volume		str. 311-6
Language		eng
Abstract		Prediction of a patient response to treatment with a drug infliximab based on oral clinical parameters is considered. As an optimal predictor the non-parametric regression is applied. Information about a specific case is provided by a basis of joint data about clinical parameters and response indicators observed on a test group of patients. The regression utilizes given clinical parameters of a selected patient to predict his response to a drug administration. In the prediction algorithm a similarity between given parameters and those in the data base is calculated. It is then used as a weight by which the response of a patient from the test group is accounted in the predicted response. The method thus resembles a prediction that is performed by a physician based upon comparison of a treated patient with previously tested ones. The prediction quality is estimated from the discrepancy between predicted and really observed response data. Prediction quality corresponding to particular clinical parameters provides for their ordering and selection of an optimal set of parameters that together yield the maximum quality. The method is demonstrated on the prediction of the clinical response of patients suffering from Crohn's disease to treatment with the drug infliximab. As given data the dental, periodontal, oral mucosa parameters and the subgingival flora detected by darkfield microscopy are utilized.
Descriptors		CROHN DISEASE PREDICTIVE VALUE OF TESTS TUMOR NECROSIS FACTOR PERIODONTAL INDEX ORAL HYGIENE INDEX DENTAL PLAQUE INDEX ORAL HEALTH