System and method for sleep apnea detection, computer device and storage medium

A computer device of a respiratory therapy system, comprising a memory and a processor;

• the memory stores computer-readable instruction a computer command executable on the processor, wherein when executing the computer command, the processor is configured to execute the computer-readable instruction so as to perform following steps of:
• acquiring a respiratory signal of a user through a contact sensor or a non-contact sensor and performing data segment division on the respiratory signal to divide the respiratory signal into n continuous first signal segments within a first preset period;
• wherein n is a positive integer;
• dividing each of the first signal segments into h second signal segments, and calculating respiratory intensities corresponding to the h second signal segments;
• wherein h is a positive integer and h>1;
• acquiring a local threshold value corresponding to each of the first signal segments and filtering out a third signal segment from each of the first signal segments;
• wherein the third signal segment is consisted of multiple second signal segments from the h second signal segments, wherein the multiple second signal segments each has a duration time longer than r seconds and a corresponding respiratory intensity being lower than the local threshold value;
• and wherein 0<r<10;
• calculating a time ratio of duration time of all third signal segments to a total time of the respiratory signal to obtain a percentage of sleep apnea, and substituting the percentage of sleep apnea into a pre-established regression model to obtain an apnea-hypopnea index corresponding to the respiratory signal;
• wherein the regression model is used for calculating the apnea-hypopnea index according to the percentage of sleep apnea;
• sending the apnea-hypopnea index to a smart phone of the user to be checked and printed;
• and
• activating respiratory treatment equipment connected to the respiratory therapy system, when detecting that the apnea-hypopnea index is higher,
• wherein the contact sensor or the non-contact sensor is a wearable device that integrates a sensing head, a microprocessor and a communication circuit,
• wherein said acquiring the local threshold value corresponding to each of the first signal segments comprises:
• performing data segment division on the respiratory signal to divide the respiratory signal into m fourth signal segments within a second preset period;
• wherein the first preset period is longer than the second preset period, and m is positive integer;
• calculating a plurality of respiratory intensities corresponding to each of the m fourth signal segments and filtering out m mediums of the respiratory intensities corresponding to the m fourth signal segments;
• calculating m respiration declination ratios corresponding to the m fourth signal segments based on the obtained m mediums of the respiratory intensities and m minimums of the plurality of respiratory intensities, and filtering out a medium of the m respiration declination ratios, and taking the medium of the m respiratory declination ratios as the whole respiratory fluctuation degree of the respiratory signal, wherein a respiratory declination ratio is a ratio of a difference between a medium and a minimum of the plurality of respiratory intensities corresponding to each of the m fourth signal segments to the medium of the plurality of respiratory intensities corresponding to each of the m fourth signal segments;
• and
• calculating the local threshold value corresponding to each of the first signal segments based on the h respiratory intensities corresponding to each of the first signal segments and the whole respiratory fluctuation degree.

A non-transitory computer-readable storage medium which stores a computer-readable instruction, that, when being executed by a processor, causes the processor to perform following steps of:

• pre-establishing a regression model;
• acquiring a respiratory signal of a user and performing data segment division on the respiratory signal to divide the respiratory signal into n continuous first signal segments within a first preset period;
• wherein n is a positive integer;
• dividing each of the first signal segments into h second signal segments, and calculating h respiratory intensities corresponding to the h second signal segments;
• wherein h is a positive integer and h>1;
• acquiring a local threshold value corresponding to each of the first signal segments and filtering out a third signal segment from each of the first signal segments;
• wherein the third signal segment is consisted of multiple second signal segments from the h second signal segments, wherein the multiple second signal segments each has a duration time longer than r seconds and a corresponding respiratory intensity being lower than the local threshold value;
• and wherein 0<r<10;
• calculating a time ratio of total duration time of all third signal segments to a total time of the respiratory signal to obtain a percentage of sleep apnea, and substituting the percentage of sleep apnea into the pre-established regression model to obtain an apnea-hypopnea index corresponding to the respiratory signal;
• wherein the regression model is used for calculating the apnea-hypopnea index according to the percentage of sleep apnea;
• sending the apnea-hypopnea index to a smart phone of the user to be checked and printed;
• and
• activating respiratory treatment equipment connected to the respiratory therapy system, when detecting that the apnea-hypopnea index is higher,
• wherein the contact sensor or the non-contact sensor is a wearable device that integrates a sensing head, a microprocessor and a communication circuit,
• wherein said acquiring the local threshold value corresponding to each of the first signal segments comprises:
• performing data segment division on the respiratory signal to divide the respiratory signal into m fourth signal segments within a second preset period;
• wherein the first preset period is longer than the second preset period, and m is positive integer;
• calculating a plurality of respiratory intensities corresponding to each of the m fourth signal segments and filtering out m mediums of the respiratory intensities corresponding to the m fourth signal segments;
• calculating m respiration declination ratios corresponding to the m fourth signal segments based on the obtained m mediums of the respiratory intensities and m minimums of the plurality of respiratory intensities, and filtering out a medium of the m respiration declination ratios, and taking the medium of the m respiratory declination ratios as the whole respiratory fluctuation degree of the respiratory signal, wherein a respiratory declination ratio is a ratio of a difference between a medium and a minimum of the plurality of respiratory intensities corresponding to each of the m fourth signal segments to the medium of the plurality of respiratory intensities corresponding to each of the m fourth signal segments;
• and
• calculating the local threshold value corresponding to each of the first signal segments based on the h respiratory intensities corresponding to each of the first signal segments and the whole respiratory fluctuation degree.