Deep belief network-based drug identification using near infrared spectroscopy

Near infrared spectroscopy(NIRS)analysis technology,combined with chemometrics,can be effectively used in quick and nondestructive analysis of quality and category.In this paper,an effective drug identification method by using deep belief network(DBN)with dropout mecha-nism(dropout-DBN)to model NIRS is introduced,in which dropout is employed to overcome the overfitting problem coming from the small sample.This paper tests proposed method under datasets of different sizes with the example of near infrared diffuse refectance spectroscopy of erythromycin ethylsuccinate drugs and other drugs,aluminum and nonaluminum packaged.Meanwhile,it gives experiments to compare the proposed method's performance with back propagation(BP)neural network,support vector machines(SVMs)and sparse denoising auto-encoder(SDAE).The results show that for both binary classification and multi-classification,dropout mechanism can improve the classification accuracy,and dropout-DBN can achieve best classification accuracy in almost all cases.SDAE is similar to dropout-DBN in the aspects of classification accuracy and algorithm stability,which are higher than that of BP neural network and SVM methods.In terms of training time,dropout-DBN model is superior to SDAE model,but inferior to BP neural network and SVM methods.Therefore,dropout-DBN can be used as a modeling tool with effective binary and multi-class classification performance on a spectrum sample set of small size.

Drug discrimination of Near Infrared spectroscopy based on the scaled convex hull classifier

Near Infrared spectroscopy(NIRS)has been widely used in the discrimination(classification)of pharmaceutical drugs.In real applications,however,the class imbalance of the drug samples,i.e.,the number of one drug sample may be much larger than the number of the other drugs,deceasesdrastically the discrimination performance of the classification models.To address this classimbalance problem,a new computational method--the scaled convex hull(SCH)-basedmaximum margin classifier is proposed in this paper.By a suitable selection of the reductionfactor of the SCHs generated by the two classes of drug samples,respectively,the maximalmargin classifier bet ween SCHs can be constructed which can obtain good classification per-formance.With an optimization of the parameters involved in the modeling by Cuckoo Search,a satisfied model is achieved for the classification of the drug.The experiments on spectra samplesproduced by a pharmaceutical company show that the proposed method is more effective androbust than the existing ones.

Classification of natural products as sources of drugs according to the biopharmaceutics drug disposition classification system(BDDCS)

Natural products(NPs) are compounds that are derived from natural sources such as plants, animals, and microisms. Therapeutics has benefited from numerous drug classes derived from natural product sources. The Biopharmaceutics Drug position Classification System(BDDCS) was proposed to serve as a basis for predicting the importance of transporters and enzymes in determining drug bioavailability and disposition. It categorizes drugs into one of four biopharmaceutical classes according to their water solubility and extent of metabolism. The present paper reviews 109 drugs from natural product sources: 29% belong to class 1(high solubility, extensive metabolism), 22% to class 2(low solubility, extensive metabolism), 40% to class 3(high solubility, poor metabolism), and 9% to class 4(low solubility, poor metabolism). Herein we evaluated the characteristics of NPs in terms of BDDCS class for all 109 drugs as wells as for subsets of NPs drugs derived from plant sources as antibiotics. In the 109 NPs drugs, we piled 32 drugs from plants, 50%(16) of total in class 1, 22%(7) in class 2 and 28%(9) in class 3, none found in class 4; Meantime, the antibiotics were found 5(16%) in class 2, 22(71%) in class 3, and 4(13%) in class 4; no drug was found in class 1. Based on this classification, we anticipate BDDCS to serve as a useful adjunct in evaluating the potential characteristics of new natural products.