Intracranial aneurysm area automatic detection system and method for CTA image

Abstract

The invention discloses an intracranial aneurysm area automatic detection system and a detection method for a CTA image. The system comprises: a data enhancement module which is used for carrying outenhancement treatment on collected CTA image data; a suspected area pre-selection module used for carrying out model training on the enhanced CTA image data and determining central coordinates of a suspected area; a fine target detection module used for cutting an original image according to the center coordinates, then carrying out model training, and marking the area where the intracranial aneurysm is located on the original image; a multi-dimensional fusion module used for cutting a three-dimensional image according to the center coordinates, then carrying out model training, marking the area where the intracranial aneurysm is located on the three-dimensional image, and marking the area where the intracranial aneurysm is located on the original image in combination with a fine target detection result; and an incremental training module used for acquiring a newly added CTA image and re-determining the area where the intracranial aneurysm is located.

Description

technical field [0001] The invention relates to the field of auxiliary medical devices, in particular to an automatic detection system and detection method for an intracranial aneurysm region for CTA images. Background technique [0002] Intracranial aneurysm is the main cause of non-traumatic subarachnoid hemorrhage. Brain cell damage and vasospasm caused by rupture and hemorrhage will lead to high disability and high mortality. Intracranial aneurysms are estimated to affect about 3% of adults and are extremely difficult to diagnose when the tumor has not ruptured. Studies have shown that the cause of intracranial aneurysm rupture is closely related to the location and size of the intracranial aneurysm. Therefore, an accurate detection method is an effective means to reduce the risk of intracranial aneurysm rupture at this stage. Traditional intracranial aneurysm detection methods rely on manual screening and manual labeling. This method not only requires high human capi...

AI Technical Summary

Problems solved by technology

[0006] Based on the above technical problems, the present invention designs and develops an automatic detection system for intracranial aneurysm regions in CTA images. The problem of too little

[0007] The present invention designs and develops a method for automatic detection of intracranial aneurysm regions on CTA images. The purpose of the present invention is to solve the problem that the area where intracranial aneurysms are located occupies an excessive proportion of the entire CTA image due to the small number of initial samples. less, which leads to the problem that the detection of intracranial aneurysms on CTA images cannot be accurately segmented

Images