Image Analysis for In-line Measurement of Multidimensional Size, Shape, and Polymorphic Transformation of l-Glutamic Acid Using Deep Learning-Based Image Segmentation and Classification

Abstract

In situ tracking of the crystallization process through image segmentation has been developed and has encountered many challenges including improvement of in situ image quality, optimization of algorithms, and increased computation efficiency. In this study, a new method based on computer vision was proposed using the state-of-the-art deep learning technology to track crystal individuals. For the model compound l-glutamic acid, two polymorphic forms with different morphologies were segmented and classified during a seeded polymorphic transformation process. Information such as counts, size, surface area, crystal size distribution, and morphology of α- and β-form crystals was extracted for the individual crystals during the process. A comparative analysis was conducted with traditional process analytical technologies such as in situ Raman and focus beam reflection measurement. Results show a high accuracy of segmentation and classification technique and a reliable tracking of crystals evolution. The image processing speed of up to 10 frames per second makes the proposed approach suitable for in situ tracking and control of crystallization and particulate processes. Our work in this study attempts to bridge the gap between the advanced imaging analysis technology that is available today and the specific needs of solution crystallization, to track, count, and measure the individual crystals.

In situ Image Processing and Data Binning Strategy for Particle Engineering Applications

“…Crystals of L ‐glutamic acid may adopt different forms and shapes , . Here, they first appeared as small needle forms, then adopted shapes that are multicrystalline (aggregates of multi‐needles), with a basic urchin‐like shape, which have a low solidity factor describing their shape (factor 0.5 to 0.7), compared to simple shapes of highly smooth and weakly convex/concave edges (maximum factor is 1).…”

Image Analysis: A Versatile Tool in the Manufacturing and Quality Control of Pharmaceutical Dosage Forms

“…This shows the pertinence of the sample size required to obtain results with certain accuracy [ 27 , 28 , 29 ]. Overlapping particles or agglomerates may cause deviance in the results, but with deep learning-based image segmentation this problem can be eliminated [ 30 ]. Image analysis is already of wide-ranging use in powder characterization.…”

Inferring kinetic dissolution of NaCl in aqueous glycol solution using a low‐cost apparatus and population balance model

“…The solid concentration in Equation 15is C NaCl,S = m NaCl, S /V in g/cm 3 , where m NaCl,L is the mass of solid NaCl in the suspension, and V is the mixture volume. The solid mass of NaCl is obtained from the PSD third order moment μ 3 , Equation (16). The electrolyte concentration is C NaCl,L = m NaCl, L /V (g/cm 3 ), in which m NaCl,L is obtained from the dynamic model, Equation 3:…”

“…Cardona et al [ 15 ] proposed a framework for extracting quantitative properties from PVM measurements (extendable to others in‐line methods) capable of mitigating some inherent issues of in‐line imaging capturing, such as out of focus objects, higher particle overlap possibility, and the challenges of characterizing a 3D particle by its 2D projection. Gao et al [ 16 ] developed a method for monitoring multidimensional size, shape, and polymorphic transformations using in situ images from a home‐designed microscopy camera using state‐of‐the‐art deep learning technology for individual particle tracking. Using an external high‐speed camera, De Anda et al [ 17 ] described a methodology for the classification of crystals polymorphism and morphology in real‐time providing estimates of their ratio in solution.…”