Label-free Estimation of Sarcomere Orientation from Brightfield Microscopy Images of Induced Pluripotent Stem Cell Derived Cardiomyocyte Nuclei

Human induced pluripotent stem cell derived cardiomyocytes (hiPSC-CMs) provide a platform for studying disease models and physiological conditions. Sarcomere structure orientation can be used to determine hiPSC-CM culture maturity. However, novel methods are needed for assessing their structure and cellular function. Brightfield microscopy enables continuous label-free measurement of cell cultures in vitro. Here, we propose evaluating sarcomere organization from the morphology and orientation of nuclei in brightfield images. We used publicly available image dataset consisting of brightfield complemented with stained nuclei and
$\alpha$
- actinin. We trained a U-Net-based network for segmenting nuclei from brightfield images (IOU of 0.72) and extracted the orientation and aspect ratio of the predicted and stained nuclei. We quantified myofibrillar orientation from
$\alpha$
-actinin and determined how it related to nuclei. The analysis revealed correlation between elongated nuclei and the orientation of the surrounding myofibrils. Our results indicate that brightfield data alone can provide estimates of cellular structures without staining. This provides the means to assess structure and maturity from repeated measurements of unstained cells, enabling high-throughput quantification of the in-vitro cardiomyocyte mechanobiology.