System and method for reducing size of physiological data
阅读说明:本技术 用于减小生理数据大小的系统和方法 (System and method for reducing size of physiological data ) 是由 G·N·加西亚莫利纳 S·普丰特纳 A·阿奎诺 于 2018-06-13 设计创作,主要内容包括:本公开涉及用于编码和/或解码脑活动信号以进行数据减小的系统和方法。在非限制性实施例中,接收与用户的第一睡眠会话相关联的第一用户数据。确定第一用户数据,以至少包括为第一数据大小的第一睡眠特征的第一实例。确定表示在第一时间间隔期间的第一实例的第一值。确定表示第一值的第一编码数据,第一编码数据为小于第一数据大小的第二数据大小。通过使用第一编码数据对第一用户数据进行编码来生成第二用户数据,以在第二用户数据中表示第一实例,并存储第二用户数据。(The present disclosure relates to systems and methods for encoding and/or decoding brain activity signals for data reduction. In a non-limiting embodiment, first user data associated with a first sleep session of a user is received. First user data is determined to include at least a first instance of a first sleep characteristic that is a first data size. A first value representing a first instance during a first time interval is determined. First encoded data representing a first value is determined, the first encoded data being of a second data size smaller than the first data size. Second user data is generated by encoding the first user data using the first encoding data to represent the first instance in the second user data, and the second user data is stored.)
1. A method for reducing a data size of user data associated with a sleep session, the method comprising:
receiving, from one or more sensors, first user data associated with a first sleep session of a user;
determining that the first user data includes at least a first instance of a first sleep characteristic, the first sleep characteristic being a first data size;
determining a first value representative of the first instance during a first time interval;
determining first encoded data representing the first value, the first encoded data being of a second data size smaller than the first data size;
generating second user data by encoding the first user data using the first encoding data to represent the first instance in the second user data; and is
Storing the second user data.
2. The method of claim 1, further comprising:
generating third user data by applying a first filter to the first user data before the first value is determined;
segmenting the third user data into a plurality of time intervals, the plurality of time intervals including the first time interval; and is
Determining that the first time interval includes the first instance.
3. The method of claim 1, further comprising:
obtaining the second user data;
identifying that the second user data is associated with the first value; and is
Generating third user data representing the first value, wherein the third user data comprises a reference version of the first sleep characteristic.
4. The method of claim 1, wherein receiving the first user data comprises receiving electroencephalography ("EEG") data, the at least one sleep characteristic comprising one of a first amount of EEG power in a delta band RMS, a second amount of EEG power in an β band RMS, a third amount of EEG power in a α band RMS, a hypnogram, a moving average related to EEG data, a depth of sleep, a timing of detected sleep waves, a number of sleep slow waves per unit time, an impedance, and a timing of detected sleep arousals.
5. The method of claim 1, further comprising:
determining a sampling rate associated with the first user data before the first value is determined;
generating third user data representing the first user data having the sampling rate applied to the first user data;
determining a first entropy rate associated with the first sleep characteristic; and is
Determining, for a second time interval of the third user data, that a first characteristic value corresponds to the first value based on the first characteristic value being less than a first sleep characteristic value and greater than a second sleep characteristic value.
6. The method of claim 1, wherein the first data size corresponds to megabytes of data and the second data size corresponds to kilobytes of data.
7. The method of claim 1, wherein the second user data comprises a three bit long code representing the first value.
8. The method of claim 1, further comprising:
determining that the first user data includes a second instance of the first sleep characteristic;
determining a second value representative of the second instance occurring during a second time interval; and is
Determining second encoded data representing the second value, the second encoded data being of a third data size smaller than the first data size, wherein generating the second user data further comprises encoding the first user data using the second encoded data to represent the second instance in the second user data.
9. The method of claim 8, further comprising:
obtaining the first encoded data and the second encoded data;
obtaining the second user data;
identifying that the first encoded data is associated with the first value occurring during the first time interval and the second encoded data is associated with the second value occurring during the second time interval; and is
Generating third user data representing the first and second values, wherein the third user data comprises a first reference version of the first value and a second reference version of the second value arranged based on a temporal ordering of the first and second time intervals.
10. A system for reducing a data size of user data associated with a sleep session, the system comprising:
one or more sensors;
a memory; and
one or more processors configured by machine-readable instructions to:
receive, from the one or more sensors, first user data associated with a first sleep session of a user;
determining that the first user data includes at least a first instance of a first sleep characteristic, the first sleep characteristic being a first data size;
determining a first value representative of the first instance during a first time interval;
determining first encoded data representing the first value, the first encoded data being of a second data size smaller than the first data size;
generating second user data by encoding the first user data using the first encoding data to represent the first instance in the second user data; and is
Storing the second user data.
11. The system of claim 10, wherein the one or more processors are further configured by the machine-readable instructions to:
generating third user data by applying a first filter to the first user data before the first value is determined;
segmenting the third user data into a plurality of time intervals, the plurality of time intervals including the first time interval; and is
Determining that the first time interval includes the first instance.
12. The system of claim 10, wherein the one or more processors are further configured by the machine-readable instructions to:
obtaining the second user data;
identifying that the second user data is associated with the first value; and is
Generating third user data representing the first value, wherein the third user data comprises a reference version of the first sleep characteristic.
13. The system of claim 10, wherein the one or more processors are further configured by the machine-readable instructions to:
determining a sampling rate associated with the first user data before the first value is determined;
generating third user data representing the first user data having the sampling rate applied to the first user data, the third user data;
determining a first entropy rate associated with the first sleep characteristic; and is
Determining, for a first time interval of the third user data, that a first characteristic value corresponds to the first value based on the first characteristic value being less than a first sleep characteristic value and greater than a second sleep characteristic value.
14. The system of claim 10, wherein the first data size corresponds to megabytes of data and the second data size corresponds to kilobytes of data.
15. A system configured to reduce a data size of user data associated with a sleep session, the system comprising:
means for receiving first user data associated with a first sleep session of a user;
means for determining that the first user data includes at least a first instance of a first sleep characteristic, the first sleep characteristic being a first data size;
means for determining a first value representative of the first instance during a first time interval;
means for determining first encoded data representing the first value, the first encoded data being of a second data size smaller than the first data size;
means for generating second user data to represent the first instance in the second user data by encoding the first user data using the first encoding data; and
means for storing the second user data.
Technical Field
The present disclosure relates to a system and method for reducing the size of physiological data, including, for example, encoding and/or decoding brain activity signals associated with sleep sessions to reduce the size of data representative of the brain activity signals.
Background
In recent years, a variety of consumer sleep technologies ("CST") have emerged on the market, some of which rely on electroencephalographic ("EEG") signals to enable individuals to monitor their own sleep. While such techniques exist to obtain an individual's EEG or other physiological signal for sleep monitoring, the substantial size of data (representative of the physiological signal) derived by typical sleep monitoring systems based on the physiological signal collected during one or more sleep sessions makes effective and efficient sleep monitoring difficult, if not impractical, to implement. These and other disadvantages exist.
Disclosure of Invention
Accordingly, one or more aspects of the present disclosure relate to a method for reducing a data size of user data associated with a sleep session. The method includes receiving, from one or more sensors, first user data associated with a first sleep session of a user. It is determined that the first user data includes at least a first instance of a first sleep characteristic, the first sleep characteristic being a first data size. A first value representing a first instance during a first time interval is determined. First encoded data representing a first value is determined, the first encoded data being of a second data size smaller than the first data size. Second user data is then generated by encoding the first user data using the first encoding data to represent the first instance in the second user data, and the second user data is stored.
Another aspect of the present disclosure relates to a system for reducing a data size of user data associated with a sleep session. The system includes one or more sensors, a memory, and one or more processors configured by machine-readable instructions stored by the memory to receive, from the one or more sensors, first user data associated with a first sleep session of a user. The one or more processors are further configured by the machine-readable instructions to determine that the first user data includes at least a first instance of a first sleep characteristic, the first sleep characteristic being a first data size. The one or more processors are further configured by the machine-readable instructions to determine a first value representative of a first instance during a first time interval. The one or more processors are further configured by the machine-readable instructions to determine first encoded data representing the first value, the first encoded data being a second data size that is smaller than the first data size. The one or more processors are further configured by the machine-readable instructions to generate second user data by encoding the first user data with the first encoding data to represent the first instance in the second user data, and to store the second user data.
Yet another aspect of the present disclosure is directed to a system for reducing a data size of user data associated with a sleep session. The system comprises: means for receiving first user data associated with a first sleep session of a user from one or more sensors; means for determining that the first user data includes at least a first instance of a first sleep characteristic, the first sleep characteristic being a first data size; means for determining a first value representative of a first instance during a first time interval; means for determining first encoded data representing a first value, the first encoded data being of a second data size smaller than the first data size; means for generating second user data to represent the first instance in the second user data by encoding the first user data using the first encoding data; and means for storing the second user data.
These and other objects, features, and characteristics of the present disclosure, as well as the methods of operation and functions of the related elements of structure and the combination of parts and economies of manufacture, will become more apparent upon consideration of the following description and the appended claims with reference to the accompanying drawings, all of which form a part of this specification, wherein like reference numerals designate corresponding parts in the various figures. It is to be expressly understood, however, that the drawings are for the purpose of illustration and description only and are not intended as a definition of the limits of the disclosure.
Drawings
Fig. 1A is a schematic illustration of an exemplary system configured to reduce the size of physiological data, in accordance with various embodiments;
fig. 1B is an illustrative diagram of an exemplary system including an exemplary user device including one or more sensors in accordance with various embodiments;
FIG. 2 is an illustrative flow diagram of a process for building an optimized dictionary in accordance with various embodiments;
FIG. 3 is an illustrative diagram of information values indicating various sleep characteristics compared to data size in accordance with various embodiments;
FIG. 4 is an illustrative diagram of a table of entropy values for various sleep features in accordance with various embodiments;
FIG. 5 is an illustrative diagram of encoded data used to determine a first value representative of a first instance of a sleep characteristic during a first time interval in accordance with various embodiments;
fig. 6 is an illustrative diagram for generating user data by encoding user data obtained from sensor(s) of a user device using encoded data, in accordance with various embodiments;
fig. 7 is an illustrative diagram of a process for decoding user data in accordance with various embodiments;
fig. 8 is an illustrative flow diagram of an exemplary process for decoding user data in accordance with various embodiments; and is
Fig. 9 is an illustrative diagram of various graphs corresponding to various sleep characteristics including both original user data and reconstructed user data, in accordance with various embodiments.
Detailed Description
As used herein, the singular forms "a", "an" and "the" include the plural forms unless the context clearly dictates otherwise. As used herein, the term "or" means "and/or" unless the context clearly dictates otherwise. As used herein, the statement that two or more parts or components are "coupled" shall mean that the parts can be joined or operated together either directly or indirectly (i.e., through one or more intermediate parts or components) whenever a link occurs. As used herein, "directly coupled" means that two elements are in direct contact with each other. As used herein, "fixedly coupled" or "fixed" means that two components are coupled so as to move while maintaining a constant orientation relative to each other.
As used herein, the word "unitary" means that the components are created as a single piece or unit. That is, a component that includes pieces that are created separately and then coupled together as a unit is not a "unitary" component or body. As used herein, the statement that two or more parts or components "engage" one another shall mean that the parts exert a force on one another either directly or through one or more intermediate parts or components. As used herein, the term "number" shall mean one or an integer greater than one (i.e., a plurality).
Directional phrases used herein, such as, but not limited to, top, bottom, left side, right side, upper, lower, front, back, and derivatives thereof, relate to the orientation of the elements shown in the drawings and are not limiting upon the claims unless expressly recited therein.
Fig. 1A is a schematic illustration of an
In an illustrative embodiment, the
In an example embodiment where one or
Various dynamics of the sleep process, such as, but not limited to, the duration of the NREM/REM cycles, and/or changes in sleep depth, can be used to define the correlation samples that sub-sample the time series of features. As an illustrative example, the raw EEG signal may be sampled at a frequency of 100Hz, but the sleep depth values may be sampled at 1 sample per minute with little loss of information. In one embodiment, the sensor(s) 108 are configured to acquire measurements at predefined time intervals. For example, the one or
In one embodiment, a dictionary of stored values is used for each sleep characteristic, wherein each value is configured to maximize the ratio of information content to size. Quantizing these values allows a region within the space of values of the corresponding sleep characteristic to be defined so that encoding can be performed in several bits of data. For example, sixteen regions can require 4 bits per sample, while eight regions can require 3 bits per sample. As an illustrative example, during a sleep session of
In one embodiment,
Further, in one embodiment, encoding/
Processor(s) 102 include any suitable processing circuitry capable of controlling the operation and functionality of
In one embodiment,
In one embodiment,
As illustrated by fig. 1A, as
In a non-limiting embodiment, the encoding/
I/
In a non-limiting embodiment, the
As described in more detail below, in one embodiment, the encoding/
Based on the particular type of sleep characteristic, a value representative of the sleep characteristic during a first time interval is determined, and first encoded data representative of the value is determined. In an illustrative embodiment, the sleep characteristic corresponds to a first data size and the first encoded data has a second size. In one embodiment, the encoded data is stored in
In one embodiment, the encoding/
As previously described, the
Fig. 1B is an illustrative diagram of a system 190 including an
In one embodiment, first sensor 108A corresponds to a forehead sensor configured to be substantially located on the forehead of
In one embodiment, in addition to one or more sleep features, each of sensors 108A and 108B is configured to capture EEG signals at a particular sampling rate. For example, the EEG and impedance of a particular channel (e.g., associated with one of sensors 108A and 108B) may monitor the EEG signal and impedance that may be captured thereby. The amount of data accumulated by such measurements occupies a large amount of memory. For example, the EEG signal plus impedance may correspond to approximately 180Kb of data per minute (e.g., 180 Kb/min). Within an eight (8) hour sleep session, this corresponds to approximately 86.4Mb of data. In one embodiment, although
FIG. 2 is an illustrative flow diagram of a
At
To perform the comparison, the user data is compared to one or more reference EEG data and a confidence score is determined that indicates the likelihood that the user data includes the particular sleep feature being analyzed. If the confidence score is greater than the threshold confidence score value, the encoding/
At operation 206, a first value representing a first instance during a first time interval is determined. The first value corresponds to a sleep-related information value associated with a particular type of sleep feature associated with the first instance. The first value is determined by the extent to which the sleep signal associated with the first instance can be reconstructed from stored information (e.g., reference EEG data stored by the
Fig. 3 is an
The information content is estimated for each feature included in the
a histogram for each of the sleep features can then be determined using the database of sleep records to estimate the corresponding entropy of the sleep features, as described by equation 1. Exemplary entropy values are shown in more detail in fig. 4.
Fig. 4 is an illustrative diagram of an exemplary table 400 of entropy values for various sleep features in accordance with various embodiments. As seen by table 400, in one embodiment, the entropy also corresponds to the number of bits per sample for each sleep characteristic. For example, the sleep characteristic impedance can have an entropy or bits per sample of 1.7.
The number of regions into which a particular sleep feature is to be quantified is based on the values identified within table 400. Using these values, quantization can be performed to indicate the value to be used for encoding each region. In one non-limiting embodiment, these values defining the encodings stored by
TABLE 1
α[uV]
β[uV]
δ[uV]
Impedance (L)
Slow wave density [ slow wave/sec ]]
2.1389
1.0534
1.8934
367.3564
0.2099
3.5367
1.3904
6.2918
497.6230
1.5587
5.1128
1.8366
9.8558
669.0929
3.1448
8.1437
2.5270
13.7817
913.3683
4.9673
12.5683
3.4515
18.4656
1299.1674
7.2324
19.0424
4.7998
24.2372
1884.4842
9.7197
31.6328
6.3910
31.8331
2549.3589
12.5771
43.7197
8.5267
43.0125
2760.7558
15.5912
As seen by Table 1, exemplary encoding values for various sleep characteristics are described.an impedance value relates to impedance in ohms by a polynomial regression equation.an encoding value corresponds to a sleep characteristic, such as α, β, δ, impedance and slow wave density, however, one of ordinary skill in the art will recognize that additional, alternative and/or fewer values may be included and the above is merely exemplary. α, β and each of the δ units correspond to a power in root mean square ("RMS") units (e.g., μ V). for example, a signal is filtered in a particular frequency band of interest (e.g., 8-12Hz for α) which is then squared on a per sample basis, and then averaged using a moving average window (e.g., 1 second for α, 1 second for β, 10 seconds for δ) which is then squared to obtain the final result listed above.
In one embodiment, if the first value representing the first instance of the sleep feature is greater than the first dictionary value, but less than or equal to the second dictionary value, then the first value representing the first dictionary value may be used to describe the first value, e.g., if the first value representing the sleep feature in the user data is greater than the first dictionary value, then the second dictionary value may be represented by binary code 000, etc. in one embodiment, the first value representing the first instance of the sleep feature is used to describe the first value, e.g., if the first value representing the first instance of the sleep feature in the user data is greater than the first dictionary value, but less than or equal to the second dictionary value, then the first value representing the first dictionary value may be used to describe the first value, e.g., if the first value representing the sleep feature in the user data is greater than the first dictionary value, but less than or equal to the second dictionary value, then the first value representing the sleep feature may be used to describe the first value of the sleep feature in place of the three-bit encoding system encoding data, and then the first value representing the sleep feature may be used to describe the three-bit encoding data.
Returning to FIG. 2,
At
At
Fig. 5 is an illustrative diagram 500 for determining encoded data representing a first value of a first instance of a sleep characteristic during a first time interval in accordance with various embodiments. As seen in
In one embodiment, the encoding/
In one embodiment, the comparison of the average value during the particular time interval being analyzed to the various quantization levels of the sleep characteristic is performed by the encode/
Fig. 6 is an illustrative diagram for generating user data 600 by encoding user data obtained from sensor(s) 108 of
For each time interval, encoded data representing values associated with instances of the sleep characteristic determined to be included in the original user is determined. For example, the encoding process 610 determines that during the time interval 604, the value of the identified instance of the sleep feature corresponds to a first quantization level. The quantization level is represented by the encoded data 602. Similarly, encoding process 620 determines that during time interval 608, the value of the identified instance of the sleep feature corresponds to a first quantization level represented by encoded data 606. In addition, the encoding process 630 determines that during the time interval 614, the value of the identified instance of the sleep feature corresponds to the first quantization level represented by the encoded data 612.
For example, the end result of each of encoding processes 610, 620, and 630 is user data 600, including each of encoded data 602, 606, and 612, which is temporally arranged according to time intervals 604, 608, and 614. In one embodiment, each of encoded data 602, 606, and 612 is appended to include temporal metadata indicating the corresponding time interval 604, 608, and 614 associated therewith. However, those of ordinary skill in the art will recognize that this is merely exemplary. For example, the user data 600 may alternatively be arranged as a string (e.g., three bits with a value of one (1) or (0)), which is then temporally divided by the number of characters used to represent the value of the feature (e.g., three characters of a three-bit code). This allows user data 600 to be "decoded," as described in more detail below with reference to fig. 7.
Fig. 7 is an illustrative diagram of a
In one embodiment,
After
Fig. 8 is an illustrative flow diagram of an exemplary process 800 for decoding user data in accordance with various embodiments. In a non-limiting embodiment, process 800 begins at operation 802. At operation 802, user data is obtained from a memory. In one embodiment, user data is obtained from the
At operation 804, user data is parsed into time intervals based on the encoded data. In one embodiment, the
At operation 806, a first value associated with the user data for each time interval is identified. For example, based on the three-bit code used to encode a particular time interval, a corresponding quantization level associated with the three-bit code may be identified. In one embodiment, the dictionary values of
At operation 808, a reference version of the sleep characteristic representing the first value is determined. For example, in one embodiment, a reconstructed version of a particular instance of a sleep characteristic is stored by the
At operation 810, additional user data representing instances of sleep characteristics encoded by the initially obtained user data during each time interval is generated. This is described in more detail below with reference to fig. 9. Therefore, by encoding the sleep session user data, it is possible to retain the same quality of information while reducing the data size of the user data. This improvement is very beneficial because the storage system is able to store much more information and better offline analysis of the user's sleep patterns and behavior is possible.
FIG. 9 is an illustrative diagram of
In each of the graphs 900-950, raw sensor data obtained by the sensor(s) 108 of the
In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" or "comprises" does not exclude the presence of elements or steps other than those listed in a claim. In the device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. In any device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.
While the description provided above provides details for the purpose of illustration based on what is currently considered to be the most practical and preferred embodiments, it is to be understood that such details are solely for that purpose and that the disclosure is not limited to the specifically disclosed embodiments, but, on the contrary, is intended to cover modifications and equivalent arrangements that are within the spirit and scope of the appended claims. For example, it is to be understood that the present disclosure contemplates that, to the extent possible, one or more features of any embodiment can be combined with one or more features of any other embodiment.
- 上一篇:一种医用注射器针头装配设备
- 下一篇:基于隐式采集的计算机交互的睡眠监测