Spine sagittal view form prediction method, device and equipment and storage medium

Abstract

The invention relates to the field of artificial intelligence, and discloses a spine sagittal view form prediction method, device and equipment and a storage medium, which are used for improving the accuracy of spine sagittal view form sequence prediction and form visual presentation. The spine sagittal configuration prediction method comprises the following steps: training a training model according to a sample image to obtain a configuration prediction model; obtaining a to-be-processed standing position X-ray sagittal view image, preprocessing the standing position X-ray sagittal view image, and marking space coordinates corresponding to the key points of the pelvis; according to the space coordinates, a pelvis key point heat map corresponding to the standing position X-ray sagittal view image is calculated; inputting the pelvis key point heat map into a form prediction model for form calculation to obtain a prediction result corresponding to the key point heat map; and carrying out local search on the prediction result to obtain space coordinates of the spine key points, and generating a spine sagittal form curve according to the space coordinates of the spine key points.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a method, device, equipment and storage medium for predicting the shape of the sagittal spine. Background technique [0002] The sagittal morphological sequence of the pelvis plays an important role in regulating the sagittal morphological sequence of the spine. The correlation between pelvic sequence parameters and spinal sequence parameters (eg, pelvic incidence angle and lumbar lordosis angle) has received extensive attention in clinical research and is gradually being applied in preoperative planning of spinal orthopedic related surgeries. Existing schemes usually model and analyze anatomical angles, and the spatial shape of the spine sequence cannot be intuitively displayed only by angle information, and the visualization effect is poor, which is not conducive to preoperative planning and program discussion related to disease diagnosis and orthopedics. [0003] The ex...

AI Technical Summary

Problems solved by technology

Existing schemes usually model and analyze anatomical angles, and the spatial shape of the spine sequence cannot be intuitively displayed only by angle information, and the visualization effect is poor, which is not conducive to preoperative planning and scheme discussions related to disease diagnosis and orthopedics

[0003] The existing scheme is limited to the established sequence angle parameters, and it is impossible to establish a relatively complete morphological sequence geometric model. The scalability is poor, which is not conducive to prospective research, and it is impossible to use the existing angle model to study new sequence angle and distance related parameters. That is, the accuracy rate of the existing scheme is low

Images