Transfer Learning Analysis for Subvisible Particle Flow Imaging of Pharmaceutical Formulations

Abstract

Subvisible particles are an ongoing problem in biotherapeutic injectable pharmaceutical formulations, and their identification is an important prerequisite for tracing them back to their source and optimizing the process. Flow imaging microscopy (FIM) is a favored imaging technique, mainly because of its ability to achieve rapid batch imaging of subvisible particles in solution with excellent imaging quality. This study used VGG16 after transfer learning to identify subvisible particle images acquired using FlowCam. We manually prepared standards for seven classes of particles, acquired the image information through FlowCam, and fed the images over 5 µm into VGG16 consisting of a convolutional base of VGG16 pre-trained with ImageNet data and a custom classifier for training. An accuracy of 97.51% was obtained for the test set data. The study also demonstrated that the recognition method using transfer learning outperforms machine learning methods based on morphological parameters in terms of accuracy, and has a significant training speed advantage over scratch-trained CNN. The combination of transfer learning and FIM images is expected to provide a general and accurate data-analysis method for identifying subvisible particles.

Deep learning image analysis models pretrained on daily objects are useful for the preliminary characterization of particulate pharmaceutical samples

“…Building such custom-made image analysis models, despite them being significantly smaller/simpler than the state-of-the-art architectures such as ResNet or EfficientNet (Chollet, 2021;Tan & Le, 2019) Discussing such complexities with computer vision specialists, we are likely to hear about transfer learning as a possible solution (Chollet, 2021;Long et al, 2022). The main idea is that models that are trained on different datasets and for different tasks might be useful, at least as a starting point in the training process, for applications in other domains.…”

“…Discussing such complexities with computer vision specialists, we are likely to hear about transfer learning as a possible solution (Chollet, 2021; Long et al, 2022). The main idea is that models that are trained on different datasets and for different tasks might be useful, at least as a starting point in the training process, for applications in other domains.…”

Screening for subvisible particles in pharmaceutical formulations – a case study

“…In recent times, flow imaging microscopy has been widely used to characterize and count particulate matter having profound application in environmental and pharmaceutical analytics (Long et al, 2022;Covari et al, 2015). It is a technology that uses dynamic imaging for direct particle detection, which enables rapid quantification of particle size and shape for tens of thousands of particles per sample.…”