Decoupling divide-and-conquer facial nerve segmentation method and device

Abstract

The invention discloses a decoupling divide-and-conquer facial nerve segmentation method and device, and the method comprises the steps: constructing a facial nerve segmentation model which comprises a feature extraction module, a coarse segmentation module and a fine segmentation module for the features of small facial nerve structure and low contrast; on the basis that the feature extraction module is used for extracting low-level features and a plurality of different levels of advanced features, using the coarse segmentation module for carrying out global search and fusion of facial nerve features on the different levels of advanced features, and then using the fine segmentation module for decoupling the fused features to obtain center main features; after edge detail features are obtained by integrating the center main features and the low-level features, performing space attention mechanism feature extraction on the center main features and the edge detail features to obtain a facial nerve segmentation map. The method improves the automatic segmentation precision and speed of facial nerves, and satisfies the requirement for path planning before robot artificial cochlea implantation.

Description

technical field [0001] The invention relates to the field of medical image processing, in particular to a facial nerve segmentation method and device for decoupling and dividing. Background technique [0002] Robotic cochlear implantation, an automated treatment to restore hearing to patients, relies on precise preoperative planning to avoid damage to critical anatomical structures. The identification of pericochlear tissue structures plays a crucial role in preoperative path planning. Among them, damage to the facial nerve may lead to permanent temporofacial nerve palsy, which is the most important tissue structure around the cochlea, and the distance between it and surrounding tissues is less than 1 mm. However, accurate segmentation of the facial nerve faces two major challenges: (1) The facial nerve structure is very small. The facial nerve only occupies a very small area in the CT image (in a 512×512-pixel whole-brain CT image, the facial nerve area only has 9-16 pixe...

AI Technical Summary

Problems solved by technology

(2) The contrast between the facial nerve and the surrounding tissue structure is low

These methods directly introduce features from the encoder to the decoder, ignore the effectiveness of feature aggregation at different levels, limit the effective utilization of features and introduce misleading features, thus causing the facial nerve to be confused with other regions

Images