Image-based cell quality assessment: Modeling of cell morphology and quality for clinical cell therapy

Abstract

In clinical tissue engineering, both safety and effectiveness are definite requirements that should be satisfied. Conventional cell biology techniques are facing limitations in the quality assurance step of cell production for clinical therapy. Image-based cell quality assessment offers a great potential, because it is the only way to non-destructively and repeatedly assess cellular phenotypes and irregularities. To effectively assess cell quality using the multiple parameters derived from time course cell imaging, machine learning models, which have been effectively used to connect biological phenomena with biological measurements in the field of bioinformatics, are promising approaches for achieving high accuracy. Here, we present the recent results of our successful cell quality modeling and discuss its possibility and considerations on further application in clinical cell therapy.