An automatic lower limb deep venous thrombosis segmentation method and system based on deep learning

Abstract

The invention discloses an automatic segmentation method and system for lower limb deep venous thrombosis based on deep learning. The method comprises the following steps: using the depth learning method for detecting and positioning a thrombotic focus area of a CE-MRI image to obtain an image containing a thrombotic focus; And carrying out thrombus focus segmentation on the image containing the thrombus focus by adopting a deep learning method. According to the invention, the CE-MRI image and the deep learning technology are combined to be applied to automatic thrombus segmentation of deep vein thrombus, a thrombus focus is automatically segmented through two times of deep learning, manual participation of a doctor is not needed, the burden of the doctor is reduced, efficiency is higher,objectivity and accuracy are better, and the consistency and repeatability of a segmentation result are better. The method can be widely applied to the field of medical image processing.

Description

Technical field [0001] The invention relates to the field of medical image processing, in particular to a method and system for automatic segmentation of deep vein thrombosis of lower limbs based on deep learning. Background technique [0002] Deep Vein Thrombosis (DVT) is a common disease that occurs in the deep veins of the lower extremities. The annual incidence is about 0.1%, and it is increasing year by year. It has become the third largest cardiovascular disease. In addition to symptoms such as lower limb swelling and pain, more than 50% of patients with DVT are prone to pulmonary embolism, with a mortality rate of more than 20%, and are called "silent killers". [0003] There are many imaging methods currently used for DVT examination, including ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), digital subtraction technology, etc. As a non-invasive examination, MRI has the advantages of good soft tissue contrast, full field of view and no radiation. C...

AI Technical Summary

Problems solved by technology

However, there are many deficiencies in this manual delineation method: 1) The MRI scan images are diverse, resulting in a large amount of time and energy for manual reading, which adds a heavy burden to doctors; (2) relying on doctors' clinical Experience, with strong subjectivity, cannot guarantee the consistency and repeatability of segmentation results

[0006] At present, there are no reports on the application of deep learning to the automatic segmentation of lower extremity DVT

Images