Parkinson's disease leg flexibility task evaluation method and system, storage medium and terminal

Abstract

The invention provides a Parkinson's disease leg flexibility task evaluation method and system, a storage medium and a terminal. The Parkinson's disease leg flexibility task evaluation method comprises the following steps: acquiring video information containing leg flexibility actions of a Parkinson's disease patient; acquiring a skeleton sequence of the Parkinson's disease patient based on the video information; constructing a joint space-time diagram and a joint motion space-time diagram based on the skeleton sequence; respectively inputting the joint space-time diagram and the joint motionspace-time diagram into a space-time residual attention network with model-driven sparse element diagram convolution, and respectively obtaining probability values under each evaluation score output by the joint flow and the joint motion flow; and for each evaluation score, calculating a probability value obtained by adding and fusing the weights of the joint flow and the joint motion flow, and selecting the evaluation score with the highest probability value as the evaluation score. According to the Parkinson's disease leg flexibility task evaluation method and system, the storage medium andthe terminal, automatic quantitative evaluation of the Parkinson's disease leg flexibility task is realized based on the deep learning technology.

Description

technical field [0001] The invention relates to the technical field of action recognition, in particular to a method and system for assessing leg flexibility tasks of Parkinson's disease, a storage medium and a terminal. Background technique [0002] Parkinson's disease (PD) is the second most common chronic neurodegenerative disease in the elderly worldwide. Parkinson's disease is expected to pose a huge social and economic burden due to an aging population. PD will lead to a gradual decline in the exercise capacity of patients. Therefore, the assessment of the exercise capacity of PD patients is an important basis for clinical diagnosis and intervention treatment. Neurologists typically use the Unified Parkinson's Disease Rating Scale (Movement Disorder Society-sponsored revision of the Unified Parkinson's Disease Rating Scale, MDS-UPDRS) to assess motor symptoms in PD patients. The leg mobility task is one of the important components of the MDS-UPDRS, which is very impo...

AI Technical Summary

Problems solved by technology

[0011] 2) In leg flexibility tasks, doctors often need to score according to the changes before and after the patient's action execution, that is, to conduct a global assessment through the long-range dependencies between time points in the video, but previous action recognition research often focused on the The selection of time periods with discriminative significance, and the contextual relationship of temporal relationships are rarely studied;

[0012] 3) The modeling of spatiotemporal relationships is an important step in video analysis, which helps to capture discriminative spatiotemporal features; the attention mechanism is often used to model spatiotemporal relationships, but only highlighting spatiotemporal features will lose static spatial structure information , and the attention weight mapping is usually obtained by adding additional spatial or temporal learnable layers, which will not only split the co-occurrence of space and time, but also bring more parameter calculation costs

[0016] Although the above studies have proposed a variety of ways to model the spatio-temporal relationship of video sequences, they often extract the discriminative features of space and time respectively, there is a spatio-temporal fragmentation, and they all need to add an additional attention mechanism, which brings unnecessary computational and storage overhead

Images