A NONPARAMETRIC REGRESSION APPROACH TO SYRINGE GRADING FOR QUALITY IMPROVEMENT

In the biomedical products industry, measures of the quality of indivi dual clinical specimens or manufacturing production units are often av ailable in the form of high-dimensional data such as continuous record ings obtained from an analytical instrument. These recordings are then examined by experts in the field who extract certain features and use these to classify individuals. To formalize and quantify this procedu re, an approach for extracting features from recordings based on nonpa rametric regression is described. These features are then used to buil d a classification model that incorporates the knowledge of the expert . The procedure is illustrated with the problem of grading of syringes from associated friction profile data. Features of the syringe fricti on profiles used in the classification are extracted via smoothing spl ines, and grades of the syringes are assigned by an expert tribologist . A nonlinear classification model is constructed to predict syringe g rades based on the extracted features. The classification model makes it possible to grade syringes automatically without expert inspection. Using leave-one-out cross-validation, the prediction accuracy of the classification model is found to be about the same as the accuracy obt ained from the expert.