Related Experiment Video
Updated: Jul 26, 2025
07:52
Author Spotlight: Enhancing Cryo-Electron Microscopy by Automated Data Collection and Analysis Techniques
Published on: December 1, 2023
1.1K
Evaluation of reproducible cryogel preparation based on automated image analysis using deep learning
Florian Behrendt1,2, Zoltán Cseresnyés3, Ruman Gerst3,4
1Laboratory of Organic Chemistry and Macromolecular Chemistry (IOMC), Friedrich Schiller University Jena, Jena, Germany.
Journal of Biomedical Materials Research. Part A
|June 22, 2023
Summary
Automated image analysis reliably measures cryogel pore sizes, offering a faster alternative to manual methods. This technique ensures consistent, reproducible cryogel structures for various applications.
Area of Science:
- Biomaterials Science
- Materials Engineering
- Polymer Chemistry
Background:
- Cryogels are porous, sponge-like materials with interconnected macropores, crucial for applications like tissue engineering scaffolds, filters, and membranes.
- Precise control over pore size and homogeneity is essential for optimizing cryogel performance in these applications.
- Current methods for pore size evaluation can be time-consuming and subjective.
Purpose of the Study:
- To develop and validate automated image analysis algorithms for rapid and accurate cryogel pore size evaluation.
- To investigate the reproducibility of cryogel pore structures prepared using different methods.
- To compare the performance of classical and deep learning-based image analysis techniques.
Main Methods:
- Preparation of poly(dimethylacrylamide-co-2-hydroxyethyl methacrylate) cryogels using conventional and adapted reactor setups.
- Acquisition of scanning electron microscopy (SEM) images of cryogel samples.
- Development of automated image analysis algorithms combining classical and deep learning approaches.
- Validation of automated methods against manual pore size determination and mercury intrusion porosimetry (MIP).
Main Results:
- Automated image analysis, particularly the deep learning approach, demonstrated high accuracy in pore size determination, with 99.4% of data showing trivial differences compared to manual methods.
- Conventional cryogel preparation using plastic syringes yielded highly reproducible morphologies and pore sizes (17–22 μm).
- An adapted reactor setup resulted in heterogeneous cryogels lacking defined pore structures.
Conclusions:
- Automated image analysis provides a reliable and efficient substitute for manual evaluation of cryogel pore sizes.
- Conventional cryogel preparation methods ensure high reproducibility of pore structure and size.
- The developed automated methods are crucial for quality control and optimization in cryogel-based technologies.
