Spine near-end boundary angle operation optimization method based on finite elements and machine learning

Abstract

The invention discloses a spine near-end boundary angle operation optimization method based on finite elements and machine learning. The spine near-end boundary angle operation optimization method comprises the following steps: s1, collecting patient information; s2, carrying out image segmentation on the preoperative CT; s3, respectively measuring the PJA angle alpha of the patient from the preoperative total spinal X-ray and the postoperative total spinal X-ray of the patient; s4, calculating the pressure F borne by the patient before and after the intervertebral disc operation under the upright condition, wherein the head end of the vertebral body is fixed at the uppermost end of the intervertebral disc; s5, carrying out static analysis; s6, establishing a machine learning model g (.) according to the patient information and the stress distribution change data obtained in the step s3; s7, training the machine learning model g (.); and s8, obtaining an optimal solution of the postoperative PJA. According to the method, the advantages of finite element stress analysis and machine learning efficient simulation are combined, and an automatic, personalized and accurate operation scheme optimization method is provided for a patient.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to an operation optimization method for the proximal junction angle of the spine based on finite elements and machine learning. Background technique [0002] Spinal fixation devices for the treatment of spinal deformities continue to evolve with the advent of new device designs, providing surgery with many options for correction. These changes have led to innovations in many surgical protocols, including the selection of proximal and distal fusion segments, the type of implant, and the choice of internal fixation device material. Most of the newly published patents are aimed at the improvement of surgical instruments, such as Chinese patent CN110916732A, a kind of protective cover for spinal surgery drag hooks, and CN110338897A, a kind of adjustable bending plate for orthopedic scoliosis surgery. However, the biomechanical effects of each component in a complex sur...

AI Technical Summary

Problems solved by technology

However, this study focused on describing the differences that may be caused by various surgical options from a statistical point of view, and did not recommend individualized surgical orthopedic options for each patient, nor did it provide specific orthopedic values ​​for the curvature of the spine. optimization suggestions

The biggest limitation of similar biomechanical research is that it is necessary to manually select the value or category range of the included variables. The process is quite cumbersome and inefficient, and it is not possible to automatically optimize the numerical value selection for a given range interval.

Images