| 摘要 |
A method for the analysis of a component of a medical sample by means of an autoanalyser, comprising the steps of:- (a)reacting the sample with a reagent system, the step of reacting the sample with the reagent system resulting in a change of a physically measurable quantity X; (b)measuring the physically measurable quantity X and determining therefrom a measured value R for a specified sample; and (c)processing said measured value R for the specified sample by means of a processing unit having a trained neural net and determining thereform an analytical result A, wherein the processing step further includes the step of using the trained neural net resulting from a neural net training, and wherein the neural net training includes:- (1)reacting a standard sample for which an analytical result A is known with the reagent system, the step of reacting the standard sample with the reagent system resulting in a physically measurable quantity X for the standard sample; (2)measuring the physically measurable quantity X at various times (ti = t1, t2,....tn) and determining therefrom several measured values Ri(ti) of the same physically measurable quantity X describing a time-dependent alteration of the measured quantity X and determining thereform at least one of a measured value R for the standard sample and a measurement result derived from several measured value R1 for the standard sample; (3)applying simultaneously to the input of the trained neural net a plurality of time-dependent input values for a specified sample, said plurality of time-dependent input values being selected from the group consisting of: (aa)several measured values Ri(ti) of the same physically measurable quantity X for a specified sample; (bb)several measurement results derived from several measured values Ri(ti) for a specified sample, and (cc)combinations thereof; (4)repeating steps (1) through (3) for a plurality of different standard samples for which the analytical result A is known, and thereby obtaining a trained neural net. |
