The invention discloses a man-
machine suitability evaluation method for a mattress, and relates to the field of
human engineering health monitoring, and the method comprises the steps: screening testers according to a BMI grading standard, and collecting the height, weight, gender, age, sleep
habit and
chronic pain medical history basic data of the testers to obtain the BMI values of the testers; inputting the BMI value of the tester into the mattress
hardness selection model to obtain the
hardness range of the tested mattress, arranging a to-be-tested mattress with selected
hardness in a constant-temperature and constant-
humidity experimental environment, and enabling the tester to lie on the to-be-tested mattress in sequence at a standard
supine position and a standard
lateral position. According to the method, the spine stiffness demand curve is predicted in real time through the LSTM neural network,
millisecond-level dynamic support adjustment is achieved, the response speed is remarkably increased compared with a traditional method, and spine compensatory bending in sleep can be prevented; and in combination with pressure-blood
oxygen coupled IRI index calculation,
microcirculation disturbance is accurately warned before tissue injury, and the problem of local
ischemia is solved through partition adjustment.