Application of Machine Learning in Classification and Recognition of Insoluble Particles in Antibody Formulations
Objective:To classify and modeling of insoluble particles in antibody injections collected by microfluidic imaging technology,in order to establish a classification and traceability analysis method for insoluble particles.Methods:In this study,four different types of images including bubble,silicone oil droplets,glass particles and protein particles generated by repeated freezing and thawing were obtained by the microflow imaging system.These images were divided into training set and test set.The characteristics of images in training set were studied by and trained in three Convolutional Neural Network(CNN)models,namely ResNet50,DenseNet201 and ShuffleNetV2,in order to establish data model.The recognition and classification performance of the model were tested by the images in the test set.Besides,the classification accuracy and speed of the three CNN models were compared with that of human eyes through application in an actual case.Results and Conclusion:The recognition accuracy of each model for the test set is above 96％,and the DenseNet201 model has better stability.Compared with human eye recognition,there is no significant difference in accuracy,but a significant acceleration in recognition speed.This study proves that CNN model can be applied in classification and traceability analysis of insoluble particles in protein preparations,so that targeted measures can be taken to reduce the potential risks and safety hazards of drugs.
NETL
NSTL
万方数据
维普
