The study highlights the growing interest in adipose tissue as an endocrine organ and its role in conditions like obesity, diabetes, cardiovascular disease, and cancer. Advanced technologies such as 3D cell cultures and Organ-on-a-Chip models are improving in vitro studies by replicating native adipose tissue environments.
Traditional methods for analyzing adipocyte biology, such as fluorescent staining and endpoint assays, are time-consuming, invasive, and cytotoxic. Emerging AI-based image analysis offers a label-free alternative for studying adipogenic differentiation kinetics and lipid droplet morphology in live-cell imaging.
The study presents IKOSA AI, a deep-learning platform used to automatically detect and measure adipose area, total cell area, and lipid droplets in brightfield images. Compared to ImageJ, IKOSA proved more accurate and reliable, particularly for identifying droplets in large clusters. The AI-driven analysis supports quality control and real-time decisions in microphysiological systems and drug screening studies.
View the full article in the journal and see how IKOSA supported the research