Crystal properties such as size, shape, or polymorphism have significant implications on drug quality and downstream unit operations such as bioavailability, shelf life, filtration, or tableting. For these reasons crystallization engineers are motivated to have a good understanding of the process kinetics and have tight control over crystallization. This work presents the development of a graphical user interface (GUI) to extract accurate crystal size dense-slurry systems along with experimental validation of the performance. The GUI performs image analysis, user-guided noise elimination, and can perform convolutional neural network training on in-line images to classify the crystals with respect to their median sizes. The validation of the results was done by using off-line characterization tools such as SEM and Malvern’s Mastersizer. A data analysis menu was added to help design future experiments based on the extracted crystallization data and process variables.

Application of image analysis and neural networks to extract information on key crystal attributes

Where thousands of chemistry professionals meet to share ideas and advance scientific and technical knowledge. Spring 2023 abstracts, posters, and presentations will launch at the end of March.

ACS Spring 2023

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.