Exascale learning on medical image data
Exascale learning in medical image data aims at improving the performance of decision support in the process of cancer detection, localisation and staging/grading to optimize and personalize treatment planning. The automated tools deliver decision support for physicians, speed up visual inspection of the tissue specimens and reduce subjectivity in the grading/staging process.
This use case tackles cancer detection and tissue classification in the latest cancer research challenges using histopathology images, such as CAMELYON and TUPAC. The PROCESS infrastructure allows researchers to develop more complex models and use larger amounts of data at a finer scale.
The access to exascale computing will consistently decrease the turn-around time of the experiments, pushing computation boundaries and consequently allowing researchers to develop models that are otherwise computationally unfeasible.