Label-Free Detection of Biochemical Changes during Cortical Organoid Maturation via Raman Spectroscopy and Machine Learning

Human cerebral organoids have become valuable tools in neurodevelopment research, holding promise for investigating neurological diseases and reducing drug development costs. However, clinical translation and large-scale production of brain organoids face challenges due to invasive methodologies such as immunohistochemistry and omics that are traditionally used for their investigation. These hinder real-time monitoring of organoids and highlight the need for a nondestructive approach to promote resource-efficient production and standardization and enable dynamic studies for drug testing and developmental monitoring. Here, we propose a label-free methodology utilizing Raman spectroscopy (RS) and machine learning to discern cortical organoid maturation stages and to observe their biochemical variations. We validated the method's robustness by analyzing both pluripotent stem cell-derived organoids and embryonic stem cell-derived organoids, revealing also significant biochemical variability between the two. This finding paves the way for the use of RS for longitudinal studies to observe dynamic changes in brain organoids, offering a promising tool for advancing our understanding of brain development and accelerating drug discovery.