Cerebral hemorrhage segmentation prediction model processing method, device and processing equipment

Abstract

The invention provides a processing method, device and processing equipment of a cerebral hemorrhage segmentation prediction model, and the method comprises the steps: providing a deep learning method of multi-output collaborative prediction, enabling the configured cerebral hemorrhage segmentation prediction model to add the segmentation output of a cerebral hemorrhage region and a skull region, and achieving the segmentation output of a skull region. According to the method, the model can learn segmentation prediction of the skull and the cerebral hemorrhage area at the same time, so that the discrimination capability of the model is enhanced, the segmentation prediction effect of the model on the cerebral hemorrhage area can be enhanced, meanwhile, data support of the skull area can be provided, and the method has better application value for clinical work.

Images