Related Experiment Video
Updated: Jun 7, 2026
10:04
A Neural Network-Based Identification of Developmentally Competent or Incompetent Mouse Fully-Grown Oocytes
Published on: March 3, 2018
7.1K
Non-invasive maturity assessment of iPSC-CMs based on optical maturity characteristics using interpretable AI
Fabian Scheurer1,2, Alexander Hammer1, Mario Schubert3
1Institute of Biomedical Engineering, TU Dresden, Fetscherstr. 29, Dresden 01307, Germany.
Computational and Structural Biotechnology Journal
|September 8, 2025
Summary
We developed an AI tool to automatically assess human induced pluripotent stem cell-derived cardiomyocyte (iPSC-CM) maturity using video analysis. This non-invasive method accurately classifies iPSC-CM maturity, improving experimental reproducibility.
Area of Science:
- Cardiovascular Research
- Stem Cell Biology
- Artificial Intelligence in Medicine
Background:
- Human induced pluripotent stem cell-derived cardiomyocytes (iPSC-CMs) are crucial for drug discovery but exhibit an immature phenotype.
- Current methods to mature iPSC-CMs are labor-intensive and can damage cells.
- Assessing iPSC-CM maturity is vital for reliable experimental outcomes.
Purpose of the Study:
- To develop a non-invasive, automated method for classifying iPSC-CM maturity.
- To utilize artificial intelligence (AI) and video-based motion analysis for maturity assessment.
- To improve the efficiency and reproducibility of iPSC-CM studies.
Main Methods:
- Collected 230 video recordings of iPSC-CMs at different maturity stages (d21 and d42 with maturation media).
- Extracted 10 motion features using Maia software and employed a support vector machine (SVM) for classification.
- Optimized SVM hyperparameters using grid search and 5-fold cross-validation.
Main Results:
- The AI model achieved high accuracy (99.5% ± 1.1%) in classifying iPSC-CM maturity on a test set.
- Shapley Additive Explanations (SHAP) identified key features: displacement, relaxation-rise time, and beating duration.
- The method demonstrated potential for reducing variability in functional assays and drug testing.
Conclusions:
- Non-invasive optical motion analysis combined with AI offers a reliable tool for assessing iPSC-CM maturity.
- This approach can be integrated before functional readouts or drug screening.
- Automated maturity assessment can enhance the reproducibility of iPSC-CM-based research.
