Machine Learning-Driven Automated Culture Reveals Similar Performance of mTeSR+ and StemFlex in Human iP11N Pluripotent Stem Cells

Human induced pluripotent stem cells (hiPSCs) are powerful tools for disease modeling and therapeutic development, though their utility remains sensitive to operator-dependent variability and media formulation. Commercial stem cell media formulations, such as mTeSR+ and StemFlex, are designed to support robust pluripotent stem cell maintenance and improve reproducibility. While both are widely used, direct comparisons are limited. We evaluated mTeSR+ and StemFlex using the CellXpress.ai automated tissue culture platform in combination with IN Carta machine-learning-based image analysis to monitor and standardize feeding, passaging, and maintenance. Wild-type iP11N hiPSCs were cultured under standardized automated conditions with growth kinetics, immunocytochemical marker expression, and RT-qPCR assessment. Automated analysis revealed no significant differences in proliferation rates between the formulations; however, longitudinal image-based quantification identified statistically significant differences (<1%) in differentiated cell area fraction across the imaging period. Immunocytochemistry demonstrated comparable expression and localization of pluripotency-associated markers, and RT-qPCR analysis confirmed similar expression of stemness factors. These findings demonstrate that both media support comparable growth and maintenance of pluripotency under standardized automated conditions on iP11N wild-type stem cells in the short-term culture conditions analyzed. Additionally, these data highlight the value of automated imaging and machine-learning-based analysis in reducing operator-dependent variability and improving consistency in hiPSC culture, supporting its application in high-throughput stem cell research.