Pharmaceutical fingerprinting in phase space. 2. Pattern recognition

The current study introduces an approach for pattern recognition of drug ma nufacturers according to their HPLC trace impurity data. This method consid ers signals in phase space and accounts for two different types of noise: a dditive and perturbative. The pharmaceutical fingerprints are estimated as mean trajectories of HPLC trace impurity data and are used as reference mod els for recognition of new data by the minimal length classifier. The chrom atographic trace organic impurity patterns collected from six different man ufacturers of L-tryptophan are analyzed as an example. The prediction abili ty of the new method tested using three different cross-validation procedur es remains about 95% even if the number of available data in the training s ets decreases by 5 times. The accuracy of prediction in phase space is supe rior compared to results calculated using a Window Preprocessing method and artificial neural networks. The difference in performance between new and previous methods becomes more significant under particular conditions that are more adequate for practical application of the method. In addition, the current approach enables simple and comprehensive interpretation of the ca lculated results.