Science and Technology Promotion Organization (JST), Ghost Cytometer: World’s First Realization of AI-Driven High-Speed Cell Morphology Sorter
The world’s first ultrahigh-speed, high-precision optical image recognition of machine learning driven type by developing high-speed, high sensitivity, simple new optical image compression measurement method, development of new multidimensional data processing method, and fusion of microfluidic technology Realized real-time cell separation technology.
Focusing on image information analysis without human intervention, we focus on the point that there is no need for images, super fast real time analysis of compression measurement signals directly by machine learning, and the idea of ”ghost cytometry method” to see “shape without looking at images” It led to breakthrough.
A young researcher established a venture company (Sink Site Co., Ltd.) and created by combining the talent and resources of industry, government and academia.
This technology is published in Science magazine and is a pioneering example of international practical application.
— Translated by TokioX’press —