Sleep posture classification method based on pseudo item recognition fusion and HMM algorithm

Abstract

The invention discloses a sleep posture classification method based on pseudo-term recognition fusion and an HMM algorithm, and the method comprises the steps: firstly collecting an original BCG signal of a patient, carrying out the preprocessing and feature extraction, and obtaining an OM feature variable; performing Group-Lasso feature screening on the OM feature variables, and performing pseudo item identification and fusion on the screened OM feature variables to obtain an observation sequence; and intercepting an observation sequence as input of the hidden Markov model through continuous translation of the window width, carrying out HMM training, obtaining a final hidden Markov model, inputting the actual observation sequence into the model, and identifying the category of the sleep posture. According to the invention, sleep posture classification is carried out by using a model combining a pseudo-term identification and fusion method of OM feature variables and an HMM algorithm, so that a model over-fitting phenomenon caused when pseudo-terms exist in the feature variables is solved, and the problem of low precision of sleep posture identification and classification by people in the past is overcome.

Description

technical field [0001] The invention relates to the technical field of sleep monitoring, in particular to a sleep posture classification method based on pseudo-item recognition fusion and HMM algorithm containing OM feature variables. Background technique [0002] Hidden Markov Model (HMM) is a statistical model used to describe a Markov process with hidden unknown parameters. The difficulty is to determine the implicit parameters of the process from the observable parameters. These parameters are then used for further analysis, such as pattern recognition. is a statistical Markov model in which the system being modeled is considered a Markov process with unobserved (hidden) states. [0003] Hidden Markov model is a kind of Markov chain. Its state cannot be observed directly, but it can be observed through the observation vector sequence. Each observation vector is expressed as various states through certain probability density distributions. Each An observation vector is...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art, to provide a sleep posture classification method based on pseudo item recognition fusion and HMM algorithm, through non-invasive data acquisition BCG signal, preliminary feature extraction through HRV, using Group-Lasso feature Screen OM feature variables, then use pseudo-item identification and fusion methods to further process feature variables, and finally recognize sleeping posture based on HMM algorithm, which can provide data support services for human sleep management and human health monitoring through posture judgment. The overfitting phenomenon of the model caused by the presence of pseudo-items in the feature variables has overcome the problem of low accuracy in the classification of sleep postures in the past

Images