Individualized multi-threshold tumble detection method and system

Abstract

The invention provides an individualized multi-threshold tumble detection method. The method includes the steps that a first acceleration data threshold set of different grouped clusters under the designated action and a second acceleration data threshold of a tested user are extracted, and an individualized acceleration tumble detection threshold is acquired; acceleration data of the tested user are collected in real time, and acceleration change parameters and the real-time included angle between the human body and the ground are calculated; whether a pre-tumble behavior happens or not is judged, if not, the step is re-executed, and if yes, the human body behavior is traced and detected, and whether the human body is in a stable state or not is detected in a preset time period; if the human body is not in the stable state, whether the human body is in the stable state or not is detected again, and if the human body is in the stable state, whether the posture of the human body is in a lying state or not is judged according to the real-time included angle between the human body and the ground; if the posture of the human body is in the lying state, it shows that the user tumbles, and a tumble alarm is output. Threshold deviation caused by individual differences of users is avoided, misinformation and report missing phenomena are reduced, tumble detection accuracy is improved, and the requirement of human body tumble detection for high accuracy is met.

Description

technical field [0001] The technical field of biomedical signal processing of the present invention relates to a fall detection method and system, in particular to a humanized multi-threshold fall detection method and system. Background technique [0002] According to statistics from the World Organization for Disease Control and Prevention, about one-third of the elderly over the age of 65 have fallen every year in the world, but the risk of falling increases with age, and about 50%-80% Elderly people aged 80 and over are at risk of falls. Fall is one of the common serious accidents among the elderly, and its occurrence is often accompanied by serious consequences, such as: the fear of falling again, fractures, and even more serious death, which not only increase the medical expenses of the family, but also aggravate the the burden on medical institutions. At present, our country has entered the aging stage, the number of elderly people living alone is gradually increasin...

AI Technical Summary

Problems solved by technology

[0005] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide a personalized multi-threshold fall detection method and system, which is used to solve the problem that the fall detection method and system in the prior art cannot accurately obtain information on human falls, and cannot It can filter out a large number of false positives and negative negatives caused by individual differences, which cannot meet the high-precision requirements of human fall detection.

Images