In the high-throughput screening processes used for chemical and genetic screens, robotized microscopes will collect a massive quantity of images of biological samples through the analysis of thousands of wells, each containing tens to several hundreds of cells (>100,000 images per day).

Efficient image acquisition systems (machinery for automated image acquisition and large storage devices) are now available to allow practical use of these images. However, human analysis is impractically slow, thus introducing the need of machine vision algorithms for the full automation of the process.

Pattern recognition and machine vision experts, who tailor general computer vision techniques to the specific needs of biological imaging, can develop these algorithms. Critical to the successful development of machine vision algorithms is the tight collaboration between biologists, engineers and computer scientists.

We plan to establish a tight collaboration between the “ISI-GenOmics - Center of Genomic Science of IIT@SEMM” Screening Unit and the IIT computational scientists for the development of machine vision algorithms. Biologists will provide datasets of biological images that represent actual biological questions and assess the performance of elaborated computational methods.