TWI462727B - System for determining whether sleep disorders suffered based on classified physiology data and method thereof - Google Patents

Description

System for classifying physiological data to determine whether there is a sleep disorder and method thereof

A system for determining sleep disorders and a method thereof, in particular, a system for classifying physiological data to determine whether there is a sleep disorder and a method thereof.

Sleep and diet are basic physiological needs, and everyone needs to sleep. Therefore, under normal circumstances, people should not lose sleep. However, in fact, the number of visits to sleep disorders clinics shows that there are not many patients with sleep disorders. Once people have long been unable to get enough rest from good sleep, they will feel very tired and easily anxious and worried, even affecting Role function and health hazard.

If people want to judge whether they have sleep disorders, the most important way is to use a variety of physiological measuring instruments (polysomnography), a number of physiological measuring instruments can measure the apnea during sleep and the number and type of respiratory dysfunction, hypoxia index And the number of times, ECG signal, nasal and nasal airflow, respiratory movements of the chest and abdomen, blood oxygen, snoring and other physiological data, and additional instruments to measure other physiological data according to the condition of the subject, therefore, using multiple physiological measurements Measuring the physiological data during sleep will greatly help the diagnosis of sleep disorders.

However, even if a plurality of physiological measuring instruments measure the presence or absence of sleep disorders, it is necessary to measure the physiological parameters of a plurality of physiological measuring instruments for sleep for a long time and multiple times. However, the volume of a plurality of physiological measuring instruments is very large. It is very expensive and complicated to operate. Therefore, many physiological measuring instruments are usually only used by medical units or research units. Therefore, the subject must personally go to the medical unit or research unit to use a plurality of physiological measuring instruments to measure the physiological state during sleep. Therefore, long-term, multiple measurements are not convenient for the subject.

In summary, it has been known in the prior art that there is a long-term and long-term measurement of physiological data during sleep to determine whether or not there is a problem of sleep disorders. Therefore, it is necessary to propose an improved technical means to solve this problem.

In view of the prior art, there is a need for long-term and multiple measurements of physiological data during sleep to determine whether or not there is a problem of sleep disorders, and the present invention discloses a system for classifying physiological data to determine whether there is a sleep disorder and a method thereof, wherein: The invention discloses a system for classifying physiological data to determine whether there is a sleep disorder, and at least includes: a data reading module for reading physiological data measured by the subject during sleep; and a feature generating module for The physiological data generates characteristic data; the data classification module is configured to classify the physiological data according to each characteristic data; the sleep disorder determining module is configured to determine whether the subject has a corresponding sleep disorder according to each classification of the physiological data.

The method for classifying physiological data disclosed in the present invention to determine whether there is a sleep disorder comprises the steps of: reading physiological data measured by the subject during sleep; generating characteristic data according to physiological data; and classifying according to each characteristic data respectively Physiological data; determine whether the subject has a corresponding sleep disorder based on the classification of the physiological data.

The system and method disclosed in the present invention are as above, and the difference from the prior art is that the present invention determines whether the subject has a corresponding sleep disorder according to the classification of the physiological data after classifying the characteristic data according to the characteristic data of the physiological data. In order to solve the problems existing in the prior art, and to achieve the technical effect of automatically determining the sleep disorder of the subject.

The features and embodiments of the present invention will be described in detail below with reference to the drawings and embodiments, which are sufficient to enable those skilled in the art to fully understand the technical means to which the present invention solves the technical problems, and The achievable effects of the present invention.

The invention can classify the physiological data according to the characteristic data generated by the physiological data after obtaining the physiological data measured by the subject during sleep, thereby judging whether the subject has a sleep disorder. Among them, the sleep disorders mentioned in the present invention include, but are not limited to, sleep-related breathing disorders such as sleep apnea or hypopnea, and symptoms such as narcolepsy which may affect sleep quality.

The physiological data mentioned in the present invention are data obtained after performing specific measurement on an individual, for example, heart rate, respiratory number, body temperature, blood pressure, limb movement, etc., and may also be as brain wave diagram, electrocardiogram, sound reflection signal, light reflection. Signal, blood oxygen concentration, carbon dioxide concentration, skin conductance, eye movement status, eyelid activity status, volume/volume of body peripheral blood vessels, etc., may even be a collection of any of the above materials, but not limited thereto.

The characteristic data proposed by the present invention may be data identified by a waveform formed by physiological data according to time, for example, amplitude 410, frequency/cycle 420, interval time 450, duration 460, etc., as shown in FIG. Show, or a collection of the above various materials.

Hereinafter, the system operation of the present invention will be described with reference to the "2A map" of the classification physiological data of the present invention to determine whether there is a system diagram of sleep disorders. As shown in FIG. 2A, the system of the present invention includes a sensor 110, a data reading module 130, a feature generation module 150, a data classification module 160, and a sleep disorder determination module 170.

The sensor 110 is responsible for measuring the physiological data of the subject during the sleep process of the subject. The sensor 110 can be an invasive or non-invasive device. When the sensor 110 is invasive, the sensor 110 will be fully or partially buried in the subject; and when the sensor 110 is non-invasive In the formula, the sensor 110 may be in contact with the subject, such as a sphygmomanometer, a thermometer, an electroencephalogram measuring device, an electrocardiogram measuring device, etc., which may be in contact with the subject, and the sensor 110 may not need to be in contact with the subject. Contact, such as a motion sensing device, a vibration sensor, etc., does not require contact with the subject.

The data reading module 130 is responsible for connecting to the sensor in a wired or wireless manner to read the physiological data measured by the sensor 110.

In fact, the sensor 110 is not an essential component of the present invention. As shown in FIG. 2B, the sensor 110 and the storage medium 120 are external components of the present invention, and the sensor 110 is in the sleep process of the subject. After measuring the physiological data of the subject, the sensor 110 stores the measured physiological data into the storage medium 120. Therefore, in the system architecture as shown in FIG. 2B, the data reading module 130 reads the physiological data measured by the sensor 110 in the storage medium 120 by wire or wirelessly, without The physiological data measured by the sensor 110 is directly obtained.

The feature generation module 150 is responsible for generating one or more feature data according to the physiological data read by the data reading module 130, and each feature data includes one or more feature values. For example, if the sensor 110 measures the breathing of the subject, so that the data reading module 130 obtains the physiological data (breathing data) of the breathing, the feature generating module 150 can read according to the data reading module 130. Each breath in the breathing data calculates one cycle (interval of two inhaled or exhaled gases) as a characteristic value. Since the respiratory data usually contains multiple breaths, the feature generation module 150 will calculate multiple times from multiple breaths. Cycles, these calculated multiple frequencies/cycles can be combined into one feature data. Similarly, the feature generation module 150 can also calculate the amplitude of the breath (inhaled or exhaled gas) from each breath in the respiratory data. Characteristic values such as volume), duration (time to start inhaling gas to ending exhaled gas), and thus combined to generate characteristic data such as amplitude and duration of respiratory data. The feature generation module 150 can process the physiological data by using the Otsu image binary algorithm to generate the waveform formed by the physiological data corresponding time, so that the feature generation module 150 can more easily generate the above various types from the generated waveform calculation. Characteristic data.

It is particularly noted that the feature data generated by the feature generation module 150 may correspond to a sleep disorder that the sleep disorder determination module 170 can determine. For example, when a certain sleep disorder can be judged by a cycle, the feature generation module 150 generates a cycle of physiological data as feature data.

In addition, the feature generation module 150 may also extract a waveform segment or a plurality of waveform segments in each cycle of the waveform formed by the physiological data corresponding time, and respectively generate the waveform segments according to the waveform segments extracted for each cycle. A feature data corresponding to the cycle, or a feature data generated by all the extracted waveform segments.

The data classification module 160 is responsible for classifying the physiological data read by the data reading module 130 according to various feature data generated by the feature generation module 150. In some embodiments, the data classification module 160 may calculate statistical values corresponding to the respective feature data according to the feature values in the feature data generated by the feature generation module 150, for example, according to all the feature values in each feature data. Calculating the median or various averages, and generating physiological data of the characteristic data according to the calculated statistical value corresponding to each characteristic data, for example, when the physiological data is respiratory data, the data classification module 160 The respiratory (physiological data) of the subject can be classified according to the statistical value of the amplitude as deep or shallow, and the respiratory (physiological data) of the subject can be judged to be fast or slow according to the statistical value of the period or the statistical value of the duration. However, the manner in which the data classification module 160 classifies the physiological data according to the feature data is not limited to the above.

The sleep disorder determination module 170 is responsible for determining whether the subject has a specific sleep disorder based on various classifications of the physiological data. In some embodiments, the sleep disorder determination module 170 may search for a sleep disorder corresponding to various categories in a pre-established correspondence table, or determine a sleep disorder for various classifications by a predetermined judgment process, but the sleep disorder judgment mode The manner in which group 170 determines sleep disorders based on various classifications of physiological data is not limited to the above.

Next, an operational system and method of the present invention will be described with reference to an embodiment, and reference is made to the flowchart of the method for determining whether or not there is a sleep disorder by referring to the "3A map" of the present invention. In the present embodiment, it is assumed that the sensor 110 measures the breathing of the subject, that is, the physiological data in this embodiment is various materials related to breathing, such as the time point of each inhalation/exhalation, gas. The volume, the composition and ratio of the gas, etc., but the invention is not limited thereto.

In the present embodiment, the sensor 110 can be included in the system of the present invention, as shown in "FIG. 2A". Thus, if the subject wants to use the present invention to perform sleep disorder detection, the subject needs to perform Sleeping, the sensor 110 can measure the physiological data of the subject during the sleep of the subject (step 301), that is, measure the respiratory data generated by the breathing of the subject during sleep, thus, the data reading mode The group 130 can directly read the respiratory data of the subject measured by the sensor 110 (step 310), thereby providing a judgment for subsequent sleep disorders.

In addition, the sensor 110 may not be included in the system of the present invention, as shown in FIG. 2B. Thus, the subject may measure the subject by the sensor 110 before detecting the sleep disorder. Breathing during sleep to generate physiological data, and storing the measured physiological data (ie, respiratory data) of the subject during sleep in the storage medium 120, thus, the subject intends to use the present invention for sleep. When the obstacle is detected, the data reading module 130 can read the respiratory data measured by the subject during sleep in the storage medium 120 outside the system of the present invention (step 310).

Regardless of whether the sensor 110 is included in the system of the present invention, after the data reading module 130 reads the physiological data measured by the subject during sleep (step 310), the feature generating module 150 may be based on the data. The physiological data read by the reading module 130 generates feature data (step 330). In this embodiment, it is assumed that the feature generation module 150 separately calculates the cycle of each breath, the volume of each exhaled or inhaled gas, and each exhalation or inhalation in the respiratory data (physiological data). Characteristic values such as duration, such that the set of cycles of each breath of the subject can constitute characteristic data representing the breathing cycle, the same, the volume of each exhaled/inhaled gas, each exhalation/inhalation The set of durations may each constitute characteristic data representing the volume of exhaled or inhaled gas and the duration of exhalation or inhalation.

In fact, the feature generation module 150 can use the process shown in "FIG. 3B" to generate a corresponding waveform according to the data on the time of exhaling/inhaling gas, the volume of exhaled/inhaled gas, and the like in the respiratory data. Each of the generated waveforms extracts a waveform segment or a plurality of waveform segments (step 332), such as extracting waveform segments close to peaks or troughs, and/or extracting waveform segments that rise or fall from a positive value to zero. And calculating, according to the extracted waveform segment, a characteristic value of each cycle of the subject, a volume of each exhaled or inhaled gas, a duration of each exhalation or inhalation (step 336), by which each characteristic value is obtained The composition is characterized by the respiratory cycle, the volume of exhaled/inhaled gas, and the duration of exhalation/inhalation.

After the feature generation module 150 generates the feature data according to the physiological data (step 330), the data classification module 160 can classify the physiological data read by the data reading module 130 according to the various feature data generated by the feature generation module 150 ( Step 350). In this embodiment, it is assumed that the data classification module 160 can display the breathing cycle, the volume of the exhaled/inhaled gas, and the exhalation/inhalation generated by the feature generation module 150 according to the flow shown in FIG. 3C. The statistic values such as the median or the average corresponding to the various feature data are respectively calculated for each feature value included in the plurality of feature data such as the duration (step 352), and then the data classification module 160 can calculate the The ratio of the statistical value of the median or the mean value corresponding to the various characteristic data to the corresponding threshold value, the physiological data is classified into two categories of fast breathing and shallow breathing (step 356), so that the classification of the physiological data is completed.

After the data classification module 160 classifies the physiological data according to the feature data (step 350), the sleep disorder determination module 170 can determine whether the subject has the characteristics and characteristics according to the combination of the classification of the physiological data by the data classification module 160. A sleep disorder corresponding to the feature data generated by the module is generated (step 370). In the present embodiment, since the physiological data is classified into two categories: fast breathing and shallow breathing, the sleep disorder determining module 170 can search for a sleep disorder that meets the two categories of fast breathing and shallow breathing in a pre-established correspondence table. If the sleep disorder determination module 170 finds a sleep disorder that meets two categories of fast breathing and shallow breathing, the sleep disorder determination module 170 determines that the subject has a sleep disorder that is found, and if the sleep disorder is judged If the module 170 does not find a sleep disorder that meets the two categories of fast breathing and shallow breathing, the sleep disorder determining module 170 can determine that the subject does not have a sleep disorder that can be detected currently. In this way, the subject can measure the physiological data in the sleep only once, and the present invention can be used to know whether or not the sleep disorder is present.

In summary, it can be seen that the difference between the present invention and the prior art is that there is a technical means for judging whether or not the subject has a corresponding sleep disorder according to the classification of the physiological data after classifying the characteristic data according to the characteristic data of the physiological data. The technical effect of the prior art that requires long-term and multiple measurements of physiological data during sleep can be determined to determine whether or not there is a sleep disorder, thereby achieving a technical effect of automatically determining the sleep disorder of the subject.

Furthermore, the method for classifying physiological data of the present invention to determine whether there is a sleep disorder can be implemented in a combination of hardware, software or a combination of hardware and software, or can be implemented in a centralized manner in a computer system or spread by different components. The decentralized implementation of several interconnected computer systems.

While the embodiments of the present invention have been described above, the above description is not intended to limit the scope of the invention. Any modification of the form and details of the practice of the present invention, which is a matter of ordinary skill in the art to which the present invention pertains, is a patent protection of the present invention. range. The scope of the invention is to be determined by the scope of the appended claims.

110. . . Sensor

120. . . Storage medium

130. . . Data reading module

150. . . Feature generation module

160. . . Data classification module

170. . . Sleep disorder judgment module

410. . . amplitude

420. . . cycle

450. . . Intervals

460. . . duration

Step 301 Measure the physiological data of the subject during sleep

Step 310: reading the physiological data measured by the subject during sleep

Step 330: generating characteristic data based on physiological data

Step 332: extracting one waveform segment or multiple waveform segments from each cycle of the physiological data

Step 336 generates feature values that are combined into feature data according to the waveform segments.

Step 350: Classification of physiological data according to each characteristic data

Step 352: calculating respective statistical values corresponding to each feature data according to the feature values included in each feature data.

Step 356 classify physiological data according to each statistical value.

Step 370: determining whether the subject has a corresponding sleep disorder according to each category of physiological data

Figure 1 is a waveform diagram of an embodiment of the present invention.

Figure 2A is a system architecture diagram for classifying physiological data of the present invention to determine whether there is a sleep disorder.

FIG. 2B is another system architecture diagram for classifying physiological data to determine whether there is a sleep disorder.

Figure 3A is a flow chart of the method for classifying physiological data to determine whether there is a sleep disorder.

FIG. 3B is a flow chart of a detailed method for generating feature data according to the present invention.

Figure 3C is a flow chart showing the detailed method of classifying physiological data according to the present invention.

Step 301 Measure the physiological data of the subject during sleep

Step 310: reading the physiological data measured by the subject during sleep

Step 330: generating characteristic data based on physiological data

Step 350: Classification of physiological data according to each characteristic data

Step 370: determining whether the subject has a corresponding sleep disorder according to each category of physiological data

Claims (8)

A method for classifying physiological data to determine whether there is a sleep disorder, the method comprising at least the steps of: reading a physiological data measured by a subject during sleep; generating at least one characteristic data according to the physiological data, each of the The feature data includes at least one feature value; each statistical value corresponding to each feature data is calculated according to each feature value included in each feature data, and the physiological data is classified according to each of the statistical values; and the physiological data is classified according to the physiological data; Each of the categories determines whether the subject has a corresponding sleep disorder. The method for determining the presence or absence of a sleep disorder according to the classification physiological data described in claim 1 of the patent scope, wherein the method further comprises measuring before the step of reading the physiological data measured by the subject during sleep. The step of the subject's physiological data during sleep. The method for determining the presence or absence of a sleep disorder according to the classification of the physiological data described in claim 1 wherein the step of generating at least one of the characteristic data according to the physiological data further comprises extracting one of each cycle of the physiological data. A waveform segment or a plurality of waveform segments, and generating, according to the waveform segments, a step of combining at least one feature value of the feature data. A system for classifying physiological data to determine whether there is a sleep disorder, the system comprising at least: a data reading module for reading a physiological data measured by a subject during sleep; a feature generating module, configured to generate at least one feature data according to the physiological data, each of the feature data includes at least one feature value; and a data classification module configured to respectively determine a plurality of feature values included in each of the feature data Calculating each statistical value corresponding to each of the characteristic data, and classifying the physiological data according to each of the statistical values; and a sleep disorder determining module, configured to determine, according to each category of the physiological data, whether the subject has a phase Corresponding to sleep disorders. The system for classifying physiological data as described in claim 4 of the patent application to determine whether there is a sleep disorder system, wherein the system further comprises a sensor for measuring the physiological data. For example, the classification of physiological data as described in claim 4 of the patent application to determine whether there is a system of sleep disorders, wherein the characteristic data is a set formed by the frequency/cycle, amplitude, interval time and duration of the physiological data. For example, the system for classifying physiological data according to item 4 of the patent application to determine whether there is a sleep disorder system, wherein the feature generating module is further configured to extract a waveform segment or a plurality of waveform segments from each cycle of the physiological data. And generating at least one feature value combined into the feature data according to the waveform segments. The system for classifying physiological data as described in claim 4 of the patent application to determine whether there is a sleep disorder system, wherein the feature generation module generates a waveform of the physiological data using an Otsu image binary algorithm.