The present application relates to the technical field of medical
image analysis, in particular to a patient post-
anesthesia pain grading evaluation method based on
artificial intelligence. The method acquires multi-time
body posture images in a complete
respiratory cycle, constructs a
trunk reference
system, defines the end of the limb as an analysis point and a preset defense position as a target point; constructs a limb scheduling consumption matrix based on the distance between the analysis point and the target point, calculates the limb wandering
confusion degree combined with the element distribution, and fuses to obtain the defense intention focus degree reflecting the order of
whole body movement; the consumption matrix is subjected to geometric cost matching to obtain a limb scheduling total consumption index, and the focus degree is combined to generate a defense interference index; finally, the grading evaluation coefficient is determined according to the numerical distribution and stability of the index in the complete
respiratory cycle, thereby objectively distinguishing the pain state and the wake-up period agitation state, and significantly improving the
automation level and reliability of post-
anesthesia pain evaluation.