The drug supervision methods based on near-infrared spectroscopy analysis are heavily dependent on the chemometrics model which characterizes the relationship between spectral data and drug categories. The preliminary application of convolution neural network in spectral analysis demonstrates excellent end-to-end prediction ability, but it is sensitive to the hyper-parameters of the network. The transformer is a deep-learning model based on self-attention mechanism that compares convolutional neural networks (CNNs) in predictive performance and has an easy-todesign model structure. Hence, a novel calibration model named SpectraTr, based on the transformer structure, is proposed and used for the qualitative analysis of drug spectrum. The experimental results of seven classes of drug and 18 classes of drug show that the proposed SpectraTr model can automatically extract features from a huge number of spectra, is not dependent on pre-processing algorithms, and is insensitive to model hyperparameters. When the ratio of the training set to test set is 8:2, the prediction accuracy of the SpectraTr model reaches 100% and 99.52%, respectively, which outperforms PLS DA, SVM, SAE, and CNN. The model is also tested on a public drug data set, and achieved classification accuracy of 96.97% without preprocessing algorithm, which is 34.85%, 28.28%, 5.05%, and 2.73% higher than PLS DA, SVM, SAE, and CNN, respectively. The research shows that the SpectraTr model performs exceptionally well in spectral analysis and is expected to be a novel deep calibration model after Autoencoder networks (AEs) and CNN.

Near infrared (NIR) spectrum analysis technology has outstanding advantages such as rapid, nondestructive, pollution-free, and is widely used in food, pharmaceutical, petrochemical, agricultural products production and testing industries. Convolutional neural network (CNN) is one of the most successful methods in big data analysis because of its powerful feature extraction and abstraction ability, and it is especially suitable for solving multi-classification problems. CNN-based transfer learning is a machine learning technique, which migrates parameters of trained model to the new one to improve the performance. The transfer learning strategy can speed up the learning e±ciency of the model instead of learning from scratch. In view of the di±culty in acquisition of drug NIR spectral data and high labeling cost, this paper proposes three simple but very effective transfer learning methods for multi-manufacturer identification of drugs based on one-dimensional CNN. Compared with the original CNN, the transfer learning method can achieve better classification performance with fewer NIR spectral data, which greatly reduces the dependence on labeled NIR spectral data. At the same time, this paper also compares and discusses three different transfer learning methods, and selects the most suitable transfer learning model for drug NIR spectral data analysis. Compared with the current popular methods, such as SVM, BP, AE and ELM, the proposed method achieves higher classification accuracy and scalability in multi-variety and multi-manufacturer NIR spectrum classification experiments.

为解决传统显微图像拼接中产生的几何畸变和错位,及特征稀少造成的正确匹配率低、时效性差等问题,提出基于区域蛙跳搜索和图像轮廓匹配的拼接算法。提取连续采集且有重叠区域的图像轮廓曲线;引入轮廓线索感知相似度和均方误差距离,计算图像轮廓曲线间的相似度或曲线离散距离,并将其作为匹配的衡量指标;在决策域内采用区域蛙跳算法更新鸣叫分贝和蛙跳策略,搜索图像轮廓最优匹配,实现图像快速精确的拼接。结果表明,所提算法不仅具有较高的拼接精度和较强的稳健性,还减小了其简化匹配策略的计算量,具有较强的时效性。

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 reflectance 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.

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, deceases drastically the discrimination performance of the classification models. To address this class imbalance problem, a new computational method — the scaled convex hull (SCH)-based maximum margin classifier is proposed in this paper. By a suitable selection of the reduction factor of the SCHs generated by the two classes of drug samples, respectively, the maximal margin classifier between SCHs can be constructed which can obtain good classification performance. 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 samples produced by a pharmaceutical company show that the proposed method is more effective and robust than the existing ones.

A rapid quantitative analytical method for three components of Lonicerae Japonicae Flos solution (Lonicera Japonica Thumb.) extracted by water was developed using near-infrared (NIR) spectroscopy and the partial least-squares (PLS) method. The NIR spectra of 81 samples collected from a production line were obtained. The concentrations of secologanic acid, chlorogenic acid and galuteolin were determined by using high-performance liquid chromatography-diode array detection as the reference method. Several pretreatment methods for the NIR spectra were used during PLS calibration. The most appropriate latent variable number of the PLS factor was selected based on the standard error of cross-validation (SECV). The performance of the final PLS models was evaluated according to SECV, standard error of prediction (SEP) and determination coefficient (R²). The compounds secologanic acid, chlorogenic acid and galuteolin had SEP values of 0.030, 0.061 and 1.668 μg/mL, respectively and R² values over 0.85. This work shows that NIR spectroscopy is a rapid and convenient method for the analysis of Lonicerae Japonicae Flos solution extracted by water. The proposed method can help in the application of process analytical technology in the pharmaceutical industry, particularly in traditional Chinese medicine injections.