CN104706318B - An analysis method and device sleep - Google Patents

An analysis method and device sleep Download PDF

Info

Publication number
CN104706318B
CN104706318B CN201310687525.2A CN201310687525A CN104706318B CN 104706318 B CN104706318 B CN 104706318B CN 201310687525 A CN201310687525 A CN 201310687525A CN 104706318 B CN104706318 B CN 104706318B
Authority
CN
China
Prior art keywords
period
monitoring
sub
activity
complexity
Prior art date
Application number
CN201310687525.2A
Other languages
Chinese (zh)
Other versions
CN104706318A (en
Inventor
徐青青
王俊艳
张志鹏
许利群
Original Assignee
中国移动通信集团公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国移动通信集团公司 filed Critical 中国移动通信集团公司
Priority to CN201310687525.2A priority Critical patent/CN104706318B/en
Publication of CN104706318A publication Critical patent/CN104706318A/en
Application granted granted Critical
Publication of CN104706318B publication Critical patent/CN104706318B/en

Links

Abstract

本发明公开了一种睡眠分析方法及装置,包括:获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据;分别基于在每个子监测时间段内采集的该待监测者的多轴加速度数据,确定该待监测者在每个子监测时间段内的活动量;并分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定该待检测者在每个子监测时间段内的活动量特征值,并确定该监测时间段内活动量特征值的动态阈值;分别将每个子监测时间段内的活动量特征值与该动态阈值进行比较,得到该待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 The present invention discloses a method and a sleep analysis apparatus, comprising: a multi-axis acceleration data acquiring person to be monitored according to a preset sampling frequency acquired in each sub-period monitoring; are acquired based on the monitoring period in each sub- multi-axis acceleration data to be monitored by determining the activity to be monitored by the monitoring time period in each sub; and monitored separately for each sub-period, the monitoring period according to the sub-sub-period monitoring period corresponding to the window comprises the amount of activity within a segment, determining the amount of activity to be detected by the characteristic values ​​in each sub monitoring time period, and determining a dynamic threshold value of the characteristic value monitoring activity period respectively; monitoring the amount of activity of each sub-time period characteristic value is compared with the dynamic threshold, to obtain a first result of the sleep analysis to be monitored by the monitoring time period in each sub-sleep or awake state. 采用本发明实施例提供的方法,提高了醒睡分类的准确率。 Using the method provided in the embodiments of the present invention improves the accuracy of the classification of sleep and wakefulness.

Description

一种睡眠分析方法及装置 An analysis method and device sleep

技术领域 FIELD

[0001] 本发明涉及信号分析领域,尤其涉及一种睡眠分析方法及装置。 [0001] The present invention relates to the field of signal analysis, particularly to a method and a sleep analysis apparatus.

背景技术 Background technique

[0002] 睡眠研宄是睡眠学和脑电图学的重要组成部分,也是当今世界上科学研究的热点之一。 [0002] sleep study based on an important part of sleep and EEG studies, is today one of the hot research in the world. 多导睡眠图监测是目前国际公认的睡眠监测的“金标准”,通过贴在待监测者身上的电极,记录检测者的血氧、心电、眼动、腿动、脑电等指标,来判断待监测者的睡眠情况。 Polysomnography is the monitoring of the internationally recognized "gold standard" of sleep monitoring, to be monitored by the person who posted the electrodes, who recorded oximetry, ECG, eye movements, leg movements, EEG and other indicators, to judge's sleep to be monitored. 但多导睡眠图监测设备造价昂贵,随着科技的进步,对睡眠监测的研宄逐渐向小型化及家庭化的方向发展。 But polysomnographic monitoring equipment is expensive, with advances in technology, to monitor sleep study based on the gradual development to small and family-oriented direction. 通过采集用户睡眠期间的多轴加速度数据,利用用户睡眠期间的加速度小于清醒时的加速度这一特点,进行醒睡分析。 Sleep analyzed by multi-axis accelerometer data collected during the user sleeping, the acceleration when the acceleration is less than the awake during the sleep users to their characteristics, awake.

[0003] 现有的技术方案中,通过加速度传感器采集待监测者的多轴加速度数据,将该多轴加速度数据分为多个子数据段,将通过对多轴加速度数据进行判断得到的在子数据段对应的时间段内的活动次数,作为该时间段内的活动量,并且当该时间段内的活动量大于一固定阈值时,确定待监测者在该时间段内为清醒状态,否则,为睡眠状态,对每个子数据段进行分析,最后得到待监测者在整个监测时间段内的睡眠分析结果。 [0003] In prior art solutions, the multi-axis acceleration data acquired by the acceleration to be monitored by the sensor, the multi-axis acceleration data into a plurality of sub-segment, the sub data is determined by performing a multi-axis acceleration data obtained the number of segments corresponding to a period of activity, as the amount of activity of the period, and the period of time when activity is greater than a fixed threshold value, it is determined by the period to be monitored is awake, otherwise, sleep was analyzed for each sub-segment, it ends up to be monitored, the monitoring of sleep throughout the period of analysis results. 还可以对该睡眠分析结果进行后续处理,比如,当待监测者在长时间的睡眠过程中出现了短暂的清醒,则将该短暂清醒的状态判为睡眠,当待监测者在长时间的清醒过程中出现了短暂的睡眠,则将该短暂睡眠的状态判为清醒。 Subsequent processing may also be a result of the sleep analysis, for example, when the person to be monitored, there was a brief time awake during sleep, then the sentence is short awake sleeping, awake when the person to be monitored in the long occurred during a brief sleep, then the short sleep state ruled awake.

[0004]但是,对于不同的待监测者,睡眠习惯各不相同,可能有的待监测者睡眠期间比较安静,而有的可能睡眠期间比较多动,这就导致基于固定阈值判断的醒睡分类准确率较低。 [0004] However, for a different person to be monitored, sleep habits are different, and some may be monitored relatively quiet period of sleep, while others are more likely to move during sleep, which leads based on a fixed threshold value to determine the classification of the sleep-wake accuracy is low.

发明内容 SUMMARY

[0005]本发明实施例提供一种睡眠分析方法及装置,用以解决现有技术中存在的基于固定阈值判断的醒睡分类准确率较低的问题。 Example [0005] The present invention provides a method and a sleep analysis means to solve based on a fixed threshold determination wake prior art sleep lower classification accuracy problems.

[0006]本发明实施例提供一种获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据,所述多轴加速度数据包括多个多轴加速度,其中,一个监测时间段包括多个子监测时间段; Embodiment [0006] The present invention provides a multi-axis acceleration data acquiring person to be monitored according to a preset sampling frequency acquired in each sub monitoring period, the multi-axis acceleration data comprises a plurality of multi-axis accelerometer, wherein a monitoring period includes monitoring a plurality of sub-time periods;

[0007]分别基于在每个子监测时间段内采集的所述待监测者的多轴加速度数据,确定所述待监测者在每个子监测时间段内的活动量; [0007] respectively based on said multi-axial acceleration data collected during the monitoring period of each sub person to be monitored, to be monitored by determining the amount of activity in each sub monitoring time period;

[0008]分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定所述待检测者在每个子监测时间段内的活动量特征值,其中,子监测时间段对应的时间段窗口包括该子监测时间段以及其前后若干个子监测时间段; [0008] monitored separately for each sub-period, the amount of the active sub-period monitoring period comprises a time period corresponding to a sub-window of the monitoring, wherein determining the amount of activity to be detected by the monitoring time period in each sub value, wherein the monitoring period corresponding to the sub-period of the sub-window includes a plurality of sub-monitoring and monitoring period before and after the period of time;

[0009]根据监测时间段内所述多个子监测时间段内的活动量特征值,确定所述监测时间段内活动量特征值的动态阈值; [0009] The amount of active monitoring period of the monitoring feature value according to the plurality of sub-time period, said monitoring determining dynamic threshold values ​​characteristic activity period;

[0010]分别将每个子监测时间段内的活动量特征值与所述动态阈值进行比较,得到所述待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 Activity characteristic values ​​[0010] were each sub monitoring time period is compared with the dynamic threshold value, obtained by the analysis to be monitored first sleep or awake state to the sleep state in each sub-period monitoring results.

[0011] 采用本发明实施例提供的方法,基于子监测时间段的活动量,以及该子监测时间段对应的时间段窗口包括的其他的子监测时间段的活动量,确定子监测时间段的活动量特征值;根据整个监测时间段内子监测时间段的活动量特征值,确定整个监测时间段内活动量特征值的动态阈值;根据动态阈值来判断待监测者在每个子监测时间段睡眠还是清醒。 [0011] The method according to an embodiment of the present invention, based on the amount of activities of other sub-monitoring period child monitoring period the amount of activity, and the period of the window of the sub monitoring period corresponding to include determining the sub-monitoring period activity characteristic value; activity period of the sub-feature values ​​based on the entire monitoring period the monitoring, determining a dynamic threshold value for the entire period of monitoring activity characteristic; dynamic threshold is determined according to the monitors to be monitored in each sub-sleep period or wide awake. 相比于现有技术,提高了醒睡分类的准确率。 Compared to the prior art, to improve the accuracy of the classification of sleep and wakefulness.

[0012] 本发明实施例还提供一种睡眠分析装置,包括: [0012] Embodiments of the present invention further provides a sleep analysis apparatus, comprising:

[0013] 数据获取单元,用于获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据,所述多轴加速度数据包括多个多轴加速度,其中,一个监测时间段包括多个子监测时间段; [0013] The data acquisition unit for acquiring data to be monitored by a multi-axis accelerometer in accordance with a predetermined sampling frequency acquired in each sub monitoring period, the multi-axis acceleration data comprises a plurality of multi-axis accelerometer, wherein a monitoring time period includes monitoring a plurality of sub-time periods;

[0014]活动量确定单元,用于分别基于在每个子监测时间段内采集的所述待监测者的多轴加速度数据,确定所述待监测者在每个子监测时间段内的活动量; [0014] The activity determination means, for each multi-axis acceleration based on the collected data for each sub-period monitoring person to be monitored, by monitoring the amount of activity is determined in each sub-period of the monitoring to be;

[0015]活动量特征值确定单元,用于分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定所述待检测者在每个子监测时间段内的活动量特征值,其中,子监测时间段对应的时间段窗口包括该子监测时间段以及其前后若干个子监测时间段; [0015] The activity characteristic value determining means for separately for each sub monitoring period, according to the amount of activity in the sub-period monitoring period comprises a time period corresponding to a sub-window of the monitoring, determining that the person to be detected in each of the wherein the monitoring sub-activity time period value, wherein the monitoring period of the sub-time period corresponding to several sub-window comprises monitoring the sub-period and the monitoring period before and after;

[0016]动态阈值确定单元,用于根据监测时间段内所述多个子监测时间段内的活动量特征值,确定所述监测时间段内活动量特征值的动态阈值; [0016] The dynamic threshold value determination unit for activity characteristic value based on the monitoring period in the time period a plurality of local monitors to determine the dynamic threshold period of time said monitoring activity characteristic value;

[0017]处理单元,用于分别将每个子监测时间段内的活动量特征值与所述动态阈值进行比较,得到所述待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 [0017] a processing unit, respectively, for each sub-period monitoring activity characteristic value is compared with the dynamic threshold value, to be obtained by monitoring the monitoring period of each sub-sleep or awake first a sleep analysis results.

[0018]本申请的其它特征和优点将在随后的说明书中阐述,并且,部分地从说明书中变得显而易见,或者通过实施本申请而了解。 [0018] Further features of the present disclosure and advantages will be set forth in the description which follows, and in part will be obvious from the description, or learned by practice of the present application. 本申请的目的和其他优点可通过在所写的说明书、权利要求书、以及附图中所特别指出的结构来实现和获得。 The objectives and other advantages of the present disclosure may be realized and attained by the written description, claims, and drawings structure particularly pointed out.

附图说明 BRIEF DESCRIPTION

[0019]附图用来提供对本发明的进一步理解,并且构成说明书的一部分,与本发明实施例一起用于解释本发明,并不构成对本发明的限制。 [0019] The accompanying drawings provide a further understanding of the present invention, and constitute part of the specification, embodiments of the present invention serve to explain the invention, not to limit the present invention. 在附图中: In the drawings:

[0020]图1为本发明实施例提供的睡眠分析方法的流程图之一; [0020] The flowchart of FIG. 1 one sleep analysis method according to an embodiment of the present invention;

[0021]图2为本发明实施例提供的睡眠分析方法的流程图之二; [0021] The flowchart of FIG. 2 embodiment bis sleep analysis method provided by the present invention;

[0022]图3为本发明实施例提供的低频多轴加速度数据进行睡眠分析的流程图; Flowchart [0022] FIG. 3 of the present invention the low frequency sleep analysis data provided by the multi-axis accelerometer embodiment;

[0023]图4为本发明实施例提供的睡眠分析装置的结构示意图。 [0023] FIG. 4 is a schematic configuration sleep analysis apparatus according to an embodiment of the present invention.

具体实施方式 Detailed ways

[0024] 为了给出提高对待监测者醒睡分类的准确率的实现方案,本发明实施例提供了一种睡眠分析方法及装置,以下结合说明书附图对本发明的优选实施例进行说明,应当理解, 此处所描述的优选实施例仅用于说明和解释本发明,并不用于限定本发明。 [0024] In order to give improved treatment of sleep-wake implementation by monitoring the accuracy of classification, the embodiments of the present invention provides a method and a sleep analysis apparatus, the following description in conjunction with the accompanying drawings of the preferred embodiments of the present invention will be described, it being understood , preferred embodiments described herein are only used to illustrate and explain the present invention and are not intended to limit the present invention. 并且在不冲突的情况下,本申请中的实施例及实施例中的特征可以相互组合。 And in the case of no conflict, embodiments and features of the embodiments of the present application can be combined with each other.

[0025] 本发明实施例提供一种睡眠分析方法,具体流程如图1所示,包括: [0025] The embodiments of the present invention provides a method of sleep analysis, the specific procedure shown in Figure 1, comprising:

[0026]步骤101、获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据,该多轴加速度数据包括多个多轴加速度,其中,一个监测时间段包括多个子监测时间段。 [0026] Step 101, the multi-axis acceleration data acquiring person to be monitored according to a preset sampling frequency acquired in each sub monitoring period, the multi-axis acceleration data comprises a plurality of multi-axis accelerometer, which comprises a plurality monitoring period child monitoring period.

[0027]步骤1〇2、分别基于在每个子监测时间段内采集的该待监测者的多轴加速度数据, 确定该待监测者在每个子监测时间段内的活动量。 [0027] Step 1〇2, respectively, based on the multi-axis acceleration data collected to be monitored by the monitoring time period in each of the sub, to be monitored is determined by the amount of activity in each sub-time period monitored.

[0028]步骤103、分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定该待检测者在每个子监测时间段内的活动量特征值, 其中,子监测时间段对应的时间段窗口包括该子监测时间段以及其前后若干个子监测时间段。 [0028] Step 103, one for each sub monitoring period, monitoring of the amount of the sub-event time period corresponding to the time period of the sub-window comprises monitoring time period, determining that the event to be detected by the monitoring time period in each sub the amount of the characteristic value, wherein the monitoring period corresponding to the sub-period of the sub-window includes a plurality of sub-monitoring and monitoring period before and after the period.

[0029] 步骤1〇4、根据监测时间段内该多个子监测时间段内的活动量特征值,确定该监测时间段内活动量特征值的动态阈值。 [0029] Step 1〇4, according to the monitoring period to monitor activity characteristic values ​​of the plurality of sub-time period, determining a dynamic threshold value of the monitored characteristic activity period.

[0030] 步骤1〇5、分别将每个子监测时间段内的活动量特征值与该动态阈值进行比较,得到该待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 [0030] Step 1〇5, respectively, each sub-period monitoring activity characteristic value is compared with the dynamic threshold, to give a first sleeping person to be monitored in the monitoring period of each sub-sleep or awake state analyze the results.

[0031]本发明实施例中,多轴加速度数据可以通过加速度传感器来进行采集,可以将对待监测者整晚的睡眠分析作为一个监测时间段,将该监测时间段分为多个子监测时间段, 对每个子监测时间段的多轴加速度数据进行分析。 [0031] The embodiments of the present invention, the multi-axis acceleration data may be collected by the acceleration sensor may be monitored by the night treat sleep analysis as a monitoring period, the monitoring period into a plurality of sub-time monitoring period, analysis of multi-axis acceleration data for each sub monitoring period.

[0032]采用本发明实施例提供的方法,基于子监测时间段的活动量,以及该子监测时间段对应的时间段窗口包括的其他的子监测时间段的活动量,确定子监测时间段的活动量特征值;根据整个监测时间段内子监测时间段的活动量特征值,确定整个监测时间段内活动量特征值的动态阈值;根据动态阈值来判断待监测者在每个子监测时间段睡眠还是清醒。 [0032] The method according to an embodiment of the present invention, based on the amount of activities of other sub-monitoring period child monitoring period the amount of activity, and the period of the window of the sub monitoring period corresponding to include determining the sub-monitoring period activity characteristic value; activity period of the sub-feature values ​​based on the entire monitoring period the monitoring, determining a dynamic threshold value for the entire period of monitoring activity characteristic; dynamic threshold is determined according to the monitors to be monitored in each sub-sleep period or wide awake. 相比于现有技术,提高了醒睡分类的准确率。 Compared to the prior art, to improve the accuracy of the classification of sleep and wakefulness.

[0033]下面结合附图,用具体实施例对本发明提供的方法及装置和相应系统进行详细描述。 [0033] below with reference to the accompanying drawings, a method and corresponding system and device provided by the invention described in detail with specific embodiments. 方法详细步骤如图2所示,包括: The method of detailed steps shown in Figure 2, comprising:

[0034]步骤201、对采集的待监测者在监测时间段内的多轴加速度数据进行带通滤波。 [0034] Step 201, to be monitored by the acquisition of the band-pass filtering of the multi-axis acceleration data monitoring period. 用户人体活动产生的多轴加速度有一个频率范围,对多轴加速度进行带通滤波,主要是为了去除干扰。 Multi-axis acceleration human activities user has a frequency range of the multi-axis acceleration bandpass filtered, mainly to remove the interference.

[0035]步骤202、将滤波后的数据,按照预设采样频率对每个子监测时间段内的多轴加速度数据进行采样,其中,子监测时间段可以设为1分钟。 [0035] 202, the data after the filtering step, according to a preset sampling frequency to sample the multi-axis acceleration data of each sub monitoring time period, wherein the sub period can be set to monitor one minute.

[0036]步骤203、基于采样后的多轴加速度数据,确定待监测者在每个子监测时间段内的活动量。 [0036] Step 203, based on the multi-axis acceleration data sample, to be monitored by determining the amount of activity in each sub-time period monitored.

[0037]其中,当该子监测时间段内采样点的多轴加速度大于预设加速度阈值时,确定该待监测者在该采样点对应的时刻是活动的,该采样点根据预设采样频率确定;将在该子监测时间段内确定的待监测者活动的总次数,确定为待监测者在该子监测时间段的活动量。 [0037] wherein, when the sub-period monitoring sampling points of the multi-axis acceleration greater than a predetermined acceleration threshold, it is determined that the person to be monitored at the timing corresponding to the sample point is active, the sample point is determined according to a preset sampling frequency ; total number of monitoring activities to be determined in the sub monitoring time period is determined to be monitored by monitoring the amount of activity in the sub-period. [0038]关于活动量的计算有多种方法,可以选用阈值法、过零法、面积法等,本实施例选用阈值法进行活动量的确定。 [0038] There are various calculation methods on the amount of activity, can be selected threshold, the zero-crossing method, area method or the like, for example, the choice of method to determine a threshold amount of activity present embodiment.

[0039]步骤204、确定该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量的均值、方差,分别作为该子监测时间段对应的时间段窗口内活动量的均值、方差,并确定该时间段窗口内活动量大于预设活动量的子监测时间段的个数。 [0039] Step 204, determining the amount of activity of the sub-period monitoring period corresponding to the period of the sub-window comprises monitoring mean, variance, respectively, as the mean time period of the sub-window of activity corresponding to the monitoring period, variance, and determines the active window within the time period greater than the number of sub predetermined monitoring period of time of activity.

[0040]时间段窗口可以设为5分钟,子监测时间段为1分钟,则该子监测时间段对应的时间段窗口包括的子监测时间段,即为当前子监测时间段前后各2分钟对应的子监测时间段以及甴則子监测时间段,确定该5个子监测时间段活动量的均值和方差。 [0040] The time window may be set to 5 minutes period, sub-1 minute monitoring period, the monitoring period of the sub-sub-period monitoring period corresponding to the window comprises, i.e. the current corresponding to each two minutes after the monitoring sub-period monitoring sub-sub-period and the monitoring period you, determining the amount of activity of five sub-period monitoring means and variances.

[0041]步骤205、对该时间段窗口内活动量的均值、方差以及该时间段窗口内活动量大于预设活动量的子监测时间段的个数,进行加权求和,得到该子监测时间段内的活动量特征值,将活动量特征值称为PS值。 [0041] Step 205, the average amount of activity within the window period, and the variance of the period of time the active window is greater than the number of child monitoring activity preset period of time, the weighted sum, to give the sub-time monitoring activity within a segment feature value, the feature value is called PS activity value.

[0042] pS值的确定还可以根据时间段窗口内当前子监测时间段的活动量、当前子监测时间段活动量的对数以及该时间段窗口内其他子监测时间段活动量的最大值、及变化量等来确定,加权系数为经验值。 Determining the [0042] pS value may also be based on the amount of the sub-period monitoring activities within this window period, the current number of sub-time period of monitoring the activity of the active window and the period of time to monitor the amount of time the other sub-maximum value, determining the amount of change and the like, the weighting factor is an empirical value.

[0043]步骤2〇6、确定监测时间段内多个子监测时间段内的多个PS值的均值和方差,分别作为监测时间段内PS值的均值和方差。 [0043] Step 2〇6, determining the mean and variance values ​​of the plurality of PS period monitoring period plurality of local monitors, respectively, as the mean and variance values ​​PS monitoring period.

[0044]步骤207、基于该监测时间段内PS值的均值和方差,确定PS值的动态阈值。 [0044] Step 207, based on the mean value and variance of the monitored time period PS, PS determining dynamic threshold value.

[0045]当该监测时间段内活动量特征值的均值与方差之和大于第一预设活动量特征值阈值时,将该第一预设特征值阈值确定为该监测时间段内活动量特征值的动态阈值; [0045] When mean and variance of the monitoring feature value period of activity and an amount greater than a first predetermined threshold value characteristic when active, wherein the first predetermined threshold value is determined for the period of time wherein the amount of monitoring activities dynamic threshold value;

[0046]当该监测时间段内活动量特征值的均值与方差之和不大于第一预设活动量特征值阈值,且该监测时间段内活动量特征值的均值与方差之和小于第二预设活动量特征值阈值时,将该第二预设活动量特征值阈值确定为该监测时间段内活动量特征值的动态阈值, 其中,该第二预设活动量特征值阈值小于该第一预设活动量特征值阈值; [0046] When mean and variance of the feature values ​​of the monitoring activity and the time period is not greater than a first predetermined threshold value characteristic activity, and the mean and variance of the monitored period of activity and less than the second eigenvalue when a predetermined threshold value characteristic activity, wherein the second predetermined activity threshold value for determining dynamic threshold values ​​characteristic activity monitoring time period, wherein the second predetermined characteristic activity is less than the first threshold value wherein a predetermined activity threshold value;

[0047]当该监测时间段内活动量特征值的均值与方差之和不大于第一预设活动量特征值阈值,且该监测时间段内活动量特征值的均值与方差之和不小于第二预设特征值阈值时,将该监测时间段内活动量特征值的均值与方差之和确定为该监测时间段内活动量特征值的动态阈值。 [0047] When mean and variance of the feature values ​​of the monitoring activity and the time period is not greater than a first predetermined threshold value characteristic activity, and the mean and variance of the monitoring period and the amount of activity is not less than the characteristic value of wherein when two preset threshold values, the mean and variance of the feature values ​​of the monitoring activity and the time period for determining dynamic threshold values ​​characteristic activity monitoring period.

[0048]步骤208、将子监测时间段内的PS值与该监测时间段内PS值的动态阈值进行比较, 得到待监测者在子监测时间段内的第一睡眠分析结果。 [0048] Step 208, the PS value of the sub monitoring time period is compared with the dynamic threshold value PS monitoring period to obtain a first analysis result of the sleep period to be monitored by the local monitors.

[0049]其中,当子监测时间段的活动量特征值大于该监测时间段内活动量特征值的动态阈值时,确定待监测者在该子监测时间段内为清醒状态; [0049] wherein, when the monitoring activity characterized in the sub-period dynamic threshold value is greater than the period of monitoring activity characteristic values, it is determined to be monitored by the monitoring period in the sub-awake state;

[0050] 当子监测时间段的活动量特征值不大于该监测时间段内活动量特征值的动态阈值时,确定待监测者在该子监测时间段内为清醒状态。 [0050] When monitoring the activity characterized in the sub-period is not greater than the dynamic threshold value characteristic activity monitoring period, it is determined to be monitored by the monitoring period in the sub-awake state.

[0051] 在上述实施例提供的方法中,关于动态阈值的确定方法还可以采用其他方法,如最大类间方差算法,是一种自适应阈值方法,基于多个阈值,针对每个阈值,将清醒和睡眠作为两个类别,计算类间方差,将使得两类的类间方差最大的阈值作为最终阈值;熵阈值法,根据不同的阈值确定每个子监测时间段为清醒或睡眠的概率,以及对应的熵值,确定能够使熵值最大的阈值;最小误差法,此方法来源于Bayes最小误差分类方法,Eb (T)是目标类(清醒)错分到背景类(睡眠)的概率,Eo⑺是背景类(睡眠)错分到目标类(清醒)的概率,总的误差概率E (T) =Eb (T) +Eo (T),使E (T)取最小值,即为最优分类方法。 [0051] In the method provided in the embodiment described above, the method of determining on a dynamic threshold value may also be employed other methods, among such as the Otsu algorithm, an adaptive thresholding method based on a plurality of threshold values ​​for each threshold value, wakefulness and sleep as two categories, between the calculated class variance will be such that between two classes largest variance threshold as the final threshold; entropy threshold, determining for each sub monitoring period probability awake or sleep according to different threshold values, and corresponding entropy, maximum entropy is possible to determine a threshold value; minimum error method, this method is derived from Bayes minimum error classification, Eb (T) is the probability of the target class (awake) assigned to the background class error (sleep) is, Eo⑺ background class (sleep) error probability assigned to the target class (clear), the overall probability of error E (T) = Eb (T) + Eo (T), so that E (T) takes the minimum value, is the optimal classification method.

[0052]另外,有些用户可能会在睡前看书、玩手机等习惯,此类活动带来的多轴加速度数据频率较低,仅采用上述处理过程有可能会被判定为睡眠状态,因此,本发明实施例还提供一种对于低频多轴加速度数据进行睡眠分析的方法,具体步骤如图3所示,包括: [0052] In addition, some users may bedtime reading, playing mobile phones and other habits, such activities bring lower multi-axis acceleration data frequency, using only the above-mentioned process is likely to be judged as a sleep state, therefore, this Example the invention also provides a method for analysis of sleep data for the low frequency multi-axis accelerometer, the specific steps shown in Figure 3, comprising:

[0053]步骤301、对监测时间段内待监测者的多轴加速度数据进行低通滤波,得到每个子监测时间段内的低频多轴加速度数据。 [0053] Step 301, the multi-axis acceleration data monitoring period to be monitored by the low-pass filtering to obtain low frequency multi-axis acceleration data of each sub-period monitoring.

[0054] 步骤302、分别确定每个子监测时间段内低频多轴加速度数据的复杂度。 [0054] Step 302, determining the complexity of each monitoring period of each sub-data of low frequency multi-axis accelerometer.

[0055] 其中,复杂度的确定方法可以有多种,本方案可以先确定子监测时间段内低频多轴加速度数据的极值个数以及相邻的极大值与极小值之差,将这两个参数进行加权求和, 确定子监测时间段内低频多轴加速度数据的复杂度。 [0055] wherein determining the complexity of the method can have a variety, the present embodiment can be monitored to determine the extreme number of sub-period multi-axis acceleration data and the low frequency difference between the maximum value and the minimum value adjacent to it, the weighted sum of these two parameters, the complexity of determining the sub-period monitoring of low frequency multi-axis acceleration data.

[0056] 步骤303、根据监测时间段内的多个子监测时间段内低频多轴加速度数据复杂度的均值和方差,确定该监测时间段内该复杂度的动态阈值。 [0056] Step 303, according to the monitoring period for monitoring a plurality of sub-time period of the low frequency complex multi-axis acceleration data mean and variance, determining a dynamic threshold value of the complexity of the monitoring period. 关于复杂度动态阈值的确定可以与上述PS值动态阈值的确定方法相同,在此不再赘述。 Determination of dynamic thresholds may be the same complexity of the above-described method for determining dynamic threshold value PS, which is not repeated herein.

[0057] 步骤304、根据子监测时间段内低频多轴加速度数据的复杂度是否大于该复杂度的动态阈值,确定待监测者在该子监测时间段内为睡眠状态或清醒状态的第二睡眠分析结果。 [0057] Step 304, depending on the complexity of the monitoring period of the low frequency sub-multi-axis acceleration data is larger than the dynamic threshold complexity, it is determined by monitoring the sub-period monitoring sleep or awake second sleep analyze the results.

[0058]当该监测时间段内复杂度的均值与方差之和大于第一预设复杂度阈值时,将该第一预设复杂度阈值确定为该监测时间段内复杂度的动态阈值; [0058] When mean and variance of the monitored period of time and complexity than the first preset threshold of complexity, the complexity of the first predetermined threshold value determining a dynamic threshold value for the monitoring period of complexity;

[0059]当该监测时间段内复杂度的均值与方差之和不大于第一预设复杂度阈值,且该监测时间段内复杂度的均值与方差之和小于第二预设复杂度阈值时,将该第二预设复杂度阈值确定为该监测时间段内复杂度的动态阈值,其中,该第二预设复杂度阈值小于该第一预设复杂度阈值; [0059] When the complexity of the monitoring period mean and variance is not greater than a first predetermined threshold complexity, and complexity of the monitoring period and the mean and variance of less than a second predetermined complexity threshold , the second predetermined complexity threshold for determining dynamic threshold complexity monitoring period, wherein the second predetermined complexity threshold is smaller than the first predetermined complexity threshold;

[0060] 当该监测时间段内复杂度的均值与方差之和不大于第一预设复杂度阈值,且该监测时间段内复杂度的均值与方差之和不小于第二预设复杂度阈值时,将该监测时间段内复杂度的均值和方差之和确定为该监测时间段内复杂度的动态阈值。 [0060] When the complexity of the monitoring period mean and variance is not greater than a first predetermined threshold complexity, and complexity of the monitoring period and the mean and variance of the complex is not less than a second predetermined threshold when, the complexity of the monitoring period and the mean and variance determining a dynamic threshold value for the monitored period of time complexity.

[0061] 步骤305、针对上述第一睡眠分析结果为睡眠状态且第二睡眠分析结果为清醒状态的子监测时间段,确定该子监测时间段的第三睡眠分析结果为清醒状态。 [0061] Step 305, the sleep analysis results for the above first to the sleep state and the second sub-monitor sleep analysis result awake period, determining a third analytical results of the sub-monitor sleep period is awake. 其中,将第三睡眠分析结果作为待监测者在监测时间段内最终的睡眠分析结果。 Among them, the third sleep as a result of the analysis to be monitored sleep eventually results in the monitoring period. 还可以基于第三睡眠分析结果对待监测者的睡眠状况进行进一步的分析。 It can also treat sleep monitoring were conducted further analysis based on the third sleep analysis results.

[0062]基于同一发明构思,根据本发明上述实施例提供的睡眠分析方法,相应地,本发明另一实施例坯提供了睡眠分析装置,装置结构示意图如图4所示,具体包括: [0062] Based on the same inventive concept, the sleep analysis process provided by the above-described embodiments of the present invention, accordingly, another embodiment of the present invention provides a blank sleep analysis apparatus, a schematic view of the device structure shown in Figure 4, comprises:

[0063]数据获取单元401,用于获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据,该多轴加速度数据包括多个多轴加速度,其中,一个监测时间段包括多个子监测时间段; [0063] The data obtaining unit 401, configured to obtain acceleration data to be monitored by a multi-axis according to a preset sampling frequency of each sub-collection monitoring period, the multi-axis acceleration data comprises a plurality of multi-axis accelerometer, wherein a monitoring time period includes monitoring a plurality of sub-time periods;

[0064]活动量确定单元4〇2,用于分别基于在每个子监测时间段内采集的该待监测者的多轴加速度数据,确定该待监测者在每个子监测时间段内的活动量; [0064] 4〇2 active amount determination unit, for each multi-axis acceleration based on the data collected to be monitored by monitoring in each sub-period, for determining the amount of activity to be monitored by the monitoring time period in each sub;

[0065] 活动量特征值确定单元403,用于分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定该待检测者在每个子监测时间段内的活动量特征值,其中,子监测时间段对应的时间段窗口包括该子监测时间段以及其前后若干个子监测时间段; [0065] The activity characteristic value determination unit 403, for monitoring separately for each sub-period, according to the amount of activity in the sub-period monitoring period comprises a time period corresponding to a sub-window of the monitoring, determining that the person to be detected in each of the wherein the monitoring sub-activity time period value, wherein the monitoring period of the sub-time period corresponding to several sub-window comprises monitoring the sub-period and the monitoring period before and after;

[0066] 动态阈值确定单元4〇4,用于根据监测时间段内该多个子监测时间段内的活动量特征值,确定该监测时间段内活动量特征值的动态阈值; [0066] 4〇4 dynamic threshold determination unit configured to monitor the amount of activity of the plurality of sub-time period according to the characteristic value monitoring time period, determining a dynamic threshold value of the monitored activity characteristic time period;

[°067]处理单元405,用于分别将每个子监测时间段内的活动量特征值与该动态阈值进行比较,得到该待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 Activity eigenvalue period of time [° 067] processing unit 405, respectively, for each sub-monitor is compared with the dynamic threshold, to give the person to be monitored in a monitoring period of each sub-sleep or awake first a sleep analysis results.

[0068]进一步的,活动量确定单元4〇2,具体用于当该子监测时间段内采样点的多轴加速度大于预设加速度阈值时,确定该待监测者在该采样点对应的时刻是活动的,该采样点根据预设采样频率进行确定;以及将在该子监测时间段内确定该待监测者活动的总次数,确定为该待监测者在该子监测时间段的活动量。 [0068] Further, determining unit 4〇2 activity, particularly when used in the monitoring sub-sampling points of the multi-axis acceleration greater than a preset period of time acceleration threshold, determining that the person to be monitored corresponding to the sample point in time is activity, the sample point is determined according to a predetermined sampling frequency; and determining the total number of the activities to be monitored by the monitoring sub-period, determined as the amount of activity to be monitored by the monitoring sub-period.

[0069]进一步的,活动量特征值确定单元403,具体用于确定该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量的均值、方差,分别作为该子监测时间段对应的时间段窗口内活动量的均值、方差; [0069] Further, the amount of activity characteristic value determination unit 403, specifically for determining the mean period of activity of the corresponding sub-period monitoring period comprises a sub-window monitoring, variance, respectively, as the sub-period monitoring the time period corresponding to the amount of the active window mean and variance;

[0070]确定该时间段窗口内活动量大于预设活动量的子监测时间段的个数; [0070] determining an activity greater than a predetermined number of monitoring the activity of the sub-segments within the time window period;

[0071]对该时间段窗口内活动量的均值、方差以及该时间段窗口内活动量大于预设活动量的子监测时间段的个数,进行加权求和,得到该子监测时间段内的活动量特征值; [0071] The time period for the average amount of the active window, and the variance of the period of time the active window is greater than a predetermined number of sub-activity monitoring period, a weighted sum, to give the sub-time period monitored activity characteristic value;

[0072]动态阈值确定单元404,具体用于:确定监测时间段内该多个子监测时间段内的多个活动量特征值的均值和方差,分别作为该监测时间段内活动量特征值的均值和方差; [0073]当该监测时间段内活动量特征值的均值与方差之和大于第一预设活动量特征值阈值时,将该第一预设特征值阈值确定为该监测时间段内活动量特征值的动态阈值; [0072] The dynamic threshold value determination unit 404, configured to: determining a period to monitor the amount of the plurality of sub-time periods of a plurality of activity monitoring feature values ​​of mean and variance, respectively, as the characteristic value of the monitoring period the mean activity and variance; when [0073] when mean and variance of the characteristic value monitoring activity over a period greater than a first predetermined threshold value characteristic activity, wherein the first pre-determined threshold value for the monitoring time period activity characteristic dynamic threshold value;

[0074]当该监测时间段内活动量特征值的均值与方差之和不大于第一预设活动量特征值阈值,且该监测时间段内活动量特征值的均值与方差之和小于第二预设活动量特征值阈值时,将该第二预设活动量特征值阈值确定为该监测时间段内活动量特征值的动态阈值, 其中,该第二预设活动量特征值阈值小于该第一预设活动量特征值阈值; [0074] When mean and variance of the feature values ​​of the monitoring activity and the time period is not greater than a first predetermined threshold value characteristic activity, and the mean and variance of the monitored period of activity and less than the second eigenvalue when a predetermined threshold value characteristic activity, wherein the second predetermined activity threshold value for determining dynamic threshold values ​​characteristic activity monitoring time period, wherein the second predetermined characteristic activity is less than the first threshold value wherein a predetermined activity threshold value;

[0075]当该监测时间段内活动量特征值的均值与方差之和不大于第一预设活动量特征值阈值,且该监测时间段内活动量特征值的均值与方差之和不小于第二预设特征值阈值时,将该监测时间段内活动量特征值的均值与方差之和确定为该监测时间段内活动量特征值的动态阈值。 [0075] When mean and variance of the feature values ​​of the monitoring activity and the time period is not greater than a first predetermined threshold value characteristic activity, and the mean and variance of the monitoring period and the amount of activity is not less than the characteristic value of wherein when two preset threshold values, the mean and variance of the feature values ​​of the monitoring activity and the time period for determining dynamic threshold values ​​characteristic activity monitoring period.

[0076]进一步的,处理单元405,具体用于当子监测时间段的活动量特征值大于该监测时间段内活动量特征值的动态阈值时,确定待监测者在该子监测时间段内为清醒状态;以及当子监测时间段的活动量特征值不大于该监测时间段内活动量特征值的动态阈值时,确定待监测者在该子监测时间段内为清醒状态。 When [0076] Further, the processing unit 405, characteristic values ​​of specific activity for the monitoring period when the sub-threshold value is greater than a dynamic time period that the monitoring activity characteristic values, which were determined to be monitored in the monitoring period is sub awake state; and wherein when the sub-activity monitoring period is not greater than the dynamic threshold time period monitoring activity characteristic values, it is determined to be monitored by the monitoring period in the sub-awake state.

[0077]进一步的,上述装置,还包括:低频数据处理单元406,用于对该多个子监测时间段内的多轴加速度数据按照预设频率进行低通滤波,得到每个子监测时间段内的低频多轴加速度数据; [0077] Further, the apparatus further comprising: a low frequency data processing unit 406, the multi-axis acceleration data for monitoring a plurality of sub-time period according to a preset low pass filtering the frequency of each sub-monitoring time period to give low frequency multi-axis acceleration data;

[0078]分别确定每个子监测时间段内低频多轴加速度数据的复杂度; [0078] were monitored to determine the complexity of each sub-period of low frequency multi-axis acceleration data;

[0079] 根据该监测时间段内的该多个子监测时间段内低频多轴加速度数据复杂度的均值和方差,确定该监测时间段内该复杂度的动态阈值; [0079] period of monitoring low complexity data mean and variance multi-axis accelerometer, determining a dynamic threshold value of the complexity of the monitoring period based on the monitoring of the plurality of sub-time period;

[0080] 根据子监测时间段内低频多轴加速度数据的复杂度是否大于该复杂度的动态阈值,确定该待监测者在该子监测时间段内为睡眠状态或清醒状态的第二睡眠分析结果; [0080] The complexity of the monitoring period of the low frequency sub-multi-axis acceleration data is greater than a dynamic threshold value of the complexity, it is determined that the person to be monitored in the sub-period monitoring sleep or awake second sleep analysis ;

[0081] 针对第一睡眠分析结果为睡眠状态且第二睡眠分析结果为清醒状态的子监测时间段,确定该子监测时间段的第三睡眠分析结果为清醒状态。 [0081] The results for the first sleep state and second sleep sleep analysis sub-period monitoring awake, sleep analysis determination result of the third sub-period monitoring is awake.

[0082] 进一步的,低频数据处理单元406,确定一个子监测时间段内低频多轴加速度数据的复杂度,具体用于确定该子监测时间段内低频多轴加速度数据的极值个数以及相邻的极大值与极小值之差;以及对该子监测时间段内低频多轴加速度数据的极值个数以及相邻的极大值与极小值之差,进行加权求和,确定子监测时间段内低频多轴加速度数据的复杂度。 [0082] Further, the low frequency data processing unit 406, determines a period of low complexity sub-monitor multi-axis acceleration data, in particular for determining the number of extreme values ​​of the sub-period monitoring low frequency and phase of the multi-axis acceleration data difference between the maximum value and the minimum value of the neighborhood; and the difference between the maximum value and the minimum value of the low frequency sub-period multi-axis acceleration monitoring data adjacent extrema number, a weighted sum is determined the complexity of the multi-axis acceleration data of the low frequency sub-period monitoring.

[0083] 进一步的,低频数据处理单元406,确定该监测时间段内复杂度的动态阈值,具体用于当该监测时间段内复杂度的均值与方差之和大于第一预设复杂度阈值时,将该第一预设复杂度阈值确定为该监测时间段内复杂度的动态阈值; [0083] Further, the low frequency data processing unit 406, determining a dynamic threshold value of the monitoring period complexity, particularly when used to monitor the mean and variance of the complexity and the time period is greater than a first predetermined threshold complexity , the first preset threshold value determination for the complexity of the dynamic threshold complexity monitoring period;

[0084] 当该监测时间段内复杂度的均值与方差之和不大于第一预设复杂度阈值,且该监测时间段内复杂度的均值与方差之和小于第二预设复杂度阈值时,将该第二预设复杂度阈值确定为该监测时间段内复杂度的动态阈值,其中,该第二预设复杂度阈值小于该第一预设复杂度阈值; [0084] When the complexity of the monitoring period mean and variance is not greater than a first predetermined threshold complexity, and complexity of the monitoring period and the mean and variance of less than a second predetermined complexity threshold , the second predetermined complexity threshold for determining dynamic threshold complexity monitoring period, wherein the second predetermined complexity threshold is smaller than the first predetermined complexity threshold;

[0085]当该监测时间段内复杂度的均值与方差之和不大于第一预设复杂度阈值,且该监测时间段内复杂度的均值与方差之和不小于第二预设复杂度阈值时,将该监测时间段内复杂度的均值和方差之和确定为该监测时间段内复杂度的动态阈值。 [0085] When the complexity of the monitoring period mean and variance is not greater than a first predetermined threshold complexity, and complexity of the monitoring period and the mean and variance of the complex is not less than a second predetermined threshold when, the complexity of the monitoring period and the mean and variance determining a dynamic threshold value for the monitored period of time complexity.

[0086] 上述各单元的功能可对应于图1至图3所示流程中的相应处理步骤,在此不再赘述。 [0086] functions of the above units may correspond to a respective processing step of FIG. 1 to FIG. 3 in the process, are not repeated here.

[0087]综上该,本发明实施例提供的方案,获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据;分别基于在每个子监测时间段内采集的该待监测者的多轴加速度数据,确定该待监测者在每个子监测时间段内的活动量;并分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定该待检测者在每个子监测时间段内的活动量特征值;以及根据监测时间段内该多个子监测时间段内的活动量特征值,确定该监测时间段内活动量特征值的动态阈值;分别将每个子监测时间段内的活动量特征值与该动态阈值进行比较,得到该待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 [0087] In summary the program provided by the embodiment of the present invention, a multi-axis acceleration data acquired according to a preset sampling frequency acquisition period of each sub-monitoring person to be monitored; were collected based on time period of each sub-monitoring the multi-axis acceleration data to be monitored by determining the activity to be monitored by monitoring the amount of each sub-time period; and monitored separately for each sub-period, the local monitors in accordance with the sub-period corresponding to the period of the monitoring window comprises amount of activity time period, determining the amount of activity to be detected by the feature value in each sub-monitoring period; and the monitoring period in accordance with the amount of activity monitoring feature values ​​of the plurality of sub-time period, it is determined that the monitoring time period activity characteristic dynamic threshold value; are each sub-period monitoring activity characteristic value is compared with the dynamic threshold, to give a first sleeping person to be monitored in the monitoring period of each sub-sleep or awake state analyze the results. 采用本发明实施例提供的方法,相比于现有技术,提局了醒睡分类的准确率。 Using the method provided in the embodiments of the present invention, compared to the prior art, mention of the Bureau of sleep and wakefulness classification accuracy.

[0088] 本申请的实施例所提供的睡眠分析装置可通过计算机程序实现。 Example sleep analysis apparatus [0088] of the present application can be achieved by providing a computer program. 本领域技术人员应该能够理解,上述的模块划分方式仅是众多模块划分方式中的一种,如果划分为其他模块或不划分模块,只要睡眠分析装置具有上述功能,都应该在本申请的保护范围之内。 Those skilled in the art should appreciate that the above-described module division are merely one of many modules in a divided manner, if other modules or divided into dividing module, should be in the scope of the present disclosure as long as the above-described sleep function analysis apparatus having within.

[0089]本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。 [0089] The present application is a method according to an embodiment of the present application, a flowchart of a computer program product and apparatus (systems) and / or described with reference to block diagrams. 应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。 It should be understood and implemented by computer program instructions and block, and the flowchart / or block diagrams each process and / or flowchart illustrations and / or block diagrams of processes and / or blocks. 可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器,使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。 These computer program instructions may be provided to a processor a general purpose computer, special purpose computer, embedded processor or other programmable data processing apparatus to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing apparatus generating in a device for implementing the flow chart or more flows and / or block diagram block or blocks in a specified functions. [0090]这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中,使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品,该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。 [0090] These computer program instructions may also be stored in a computer can direct a computer or other programmable data processing apparatus to function in a particular manner readable memory produce an article of manufacture such that the storage instruction means comprises a memory in the computer-readable instructions the instruction means implemented in a flowchart or more flows and / or block diagram block or blocks in a specified function.

[0091]这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上,使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理,从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。 [0091] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps on the computer or other programmable apparatus to produce a computer implemented so that the computer or other programmable apparatus execute instructions to provide processes for implementing a process or flows and / or block diagram block or blocks a function specified step.

[0092]显然,本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。 [0092] Obviously, those skilled in the art can make various modifications and variations to the invention without departing from the spirit and scope of the invention. 这样,倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内,则本发明也意图包含这些改动和变型在内。 Thus, if these modifications and variations of the present invention fall within the claims of the invention and the scope of equivalents thereof, the present invention intends to include these modifications and variations.

Claims (14)

1.一种睡眠分析方法,其特征在于,包括: 获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据,所述多轴加速度数据包括多个多轴加速度,其中,一个监测时间段包括多个子监测时间段; 分别基于在每个子监测时间段内采集的所述待监测者的多轴加速度数据,确定所述待监测者在每个子监测时间段内的活动量; 分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定所述待检测者在每个子监测时间段内的活动量特征值,具体包括:确定该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量的均值、方差,分别作为该子监测时间段对应的时间段窗口内活动量的均值、方差;确定该时间段窗口内活动量大于预设活动量的子监测时间段的个数;对该时间段窗口内活动 A sleep analysis method comprising: acquiring data to be monitored by a multi-axis accelerometer in accordance with a predetermined sampling frequency acquired in each sub monitoring period, the multi-axis acceleration data comprises a plurality of multi-axis accelerometer wherein a monitoring period comprises monitoring a plurality of sub-time periods, respectively; multi-axis acceleration based on the collected data in each sub-period monitoring person to be monitored, to be monitored is determined by monitoring the time period in each sub- activity; monitored separately for each sub-period, the amount of the active sub-period monitoring period comprises a time period corresponding to a sub-window of the monitoring, wherein determining the amount of activity to be detected by the monitoring time period in each sub value comprises: determining the mean period of activity of the corresponding sub-period monitoring period comprises a sub-window monitoring, variance, respectively, as the mean time period of the sub-window of activity corresponding to the monitoring period, the variance ; determining the period of time the active window is greater than the number of child monitoring a preset period of activity; the active window within the time period 的均值、方差以及该时间段窗口内活动量大于预设活动量的子监测时间段的个数,进行加权求和,得到该子监测时间段内的活动量特征值;其中,子监测时间段对应的时间段窗口包括该子监测时间段以及其前后若干个子监测时间段; 根据监测时间段内所述多个子监测时间段内的活动量特征值,确定所述监测时间段内活动量特征值的动态阈值; 分别将每个子监测时间段内的活动量特征值与所述动态阈值进行比较,得到所述待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 Mean, variance and within the period of time the active window is greater than a predetermined number of sub-activity monitoring period, a weighted sum, wherein the amount of activity to give the sub-period monitoring value; wherein, the monitoring sub-period time period corresponding to several sub-window comprises monitoring the sub-period and the monitoring period before and after; the amount of active monitoring period of the monitoring time period plurality of sub-feature value, said monitoring period is determined eigenvalues ​​activity the dynamic threshold; respectively feature value for each sub-activity monitoring time period with the dynamic threshold value, to be obtained by monitoring the monitoring period of each sub-sleep or awake first sleep analysis .
2.如权利要求1所述的方法,其特征在于,基于在子监测时间段内采集的所述待监测者的多轴加速度数据,确定所述待监测者在该子监测时间段的活动量,具体包括: 当该子监测时间段内采样点的多轴加速度大于预设加速度阈值时,确定所述待监测者在该采样点对应的时刻是活动的,所述采样点根据预设采样频率进行确定; 将在该子监测时间段内确定所述待监测者活动的总次数,确定为所述待监测者在该子监测时间段的活动量。 2. The method according to claim 1, wherein the multi-axis acceleration data based on the monitoring period in the sub-collection of the monitors to be determined by monitoring the amount of activity in the sub-period of the monitoring to be comprises: monitoring sampling points when the sub-period multi-axial acceleration is greater than a predetermined acceleration threshold, it is determined to be monitored by the timing corresponding to the sample point is active, the sampling frequency according to a preset sampling points determination; and determining the total number of the monitors to be active, to be monitored is determined by monitoring the activity in the sub-period in the sub monitoring period.
3.如权利要求1所述的方法,其特征在于,根据监测时间段内所述多个子监测时间段内的活动量特征值,确定所述监测时间段内活动量特征值的动态阈值,具体包括: 确定监测时间段内所述多个子监测时间段内的多个活动量特征值的均值和方差,分别作为所述监测时间段内活动量特征值的均值和方差; 当所述监测时间段内活动量特征值的均值与方差之和大于第一预设活动量特征值阈值时,将所述第一预设特征值阈值确定为所述监测时间段内活动量特征值的动态阈值; 当所述监测时间段内活动量特征值的均值与方差之和不大于第一预设活动量特征值阈值,且所述监测时间段内活动量特征值的均值与方差之和小于第二预设活动量特征值阈值时,将所述第二预设活动量特征值阈值确定为所述监测时间段内活动量特征值的动态阈值,其中,所述第二预设活 3. The method according to claim 1, characterized in that the monitoring time period according to an amount of active periods of the plurality of sub-monitoring feature value, determining a dynamic threshold value characteristic of the monitored period of activity, particularly comprising: determining the mean and variance value of the characteristic amounts of the plurality of sub monitoring period monitoring periods a plurality of activities, respectively, as mean and variance of the feature values ​​of the activity monitoring time period; when the monitoring period when the mean and variance of the feature values ​​of the activity and the amount of activity than the first preset threshold value characteristic, the first predetermined threshold value is determined as the characteristic value of the dynamic threshold period of time to monitor activity characteristic value; when mean and variance of the characteristic value monitoring activity over a period of not greater than a first predetermined threshold value characteristic activity, and the mean and variance of the characteristic value monitoring activity than the second preset period of time and when the value of the threshold characteristic activity, the second predetermined threshold amount of activity is determined as the characteristic value of the dynamic threshold time period the monitoring activity characteristic value, wherein the second predetermined live 量特征值阈值小于所述第一预设活动量特征值阈值; 当所述监测时间段内活动量特征值的均值与方差之和不大于第一预设活动量特征值阈值,且所述监测时间段内活动量特征值的均值与方差之和不小于第二预设特征值阈值时,将所述监测时间段内活动量特征值的均值与方差之和确定为所述监测时间段内活动量特征值的动态阈值。 Wherein the threshold amount is smaller than the first predetermined threshold value characteristic activity; when the mean and variance of the monitoring feature value period of activity is not greater than a first predetermined threshold value characteristic activity, and said monitoring when the mean and variance of the feature values ​​of activity and the time period is not less than a second predetermined threshold value characteristic, the mean and variance of the monitored period of time and activity is determined as the characteristic value of said monitoring period activity dynamic threshold value characteristic amount.
4.如权利要求1所述的方法,其特征在于,将子监测时间段内的活动量特征值与所述监测时间段内活动量特征值的动态阈值进行比较,得到待监测者在子监测时间段内的第一睡眠分析结果,具体包括: 当子监测时间段的活动量特征值大于所述监测时间段内活动量特征值的动态阈值时, 确定待监测者在该子监测时间段内为清醒状态。 4. The method according to claim 1, wherein the amount of the active period of the sub-monitor feature value and said dynamic threshold value characteristic activity monitoring time period are compared to obtain the sub to be monitored by monitoring the results of the first sleep period comprises: when monitoring the activity characterized in the sub-period dynamic threshold value is greater than the period of monitoring activity characteristic value, it is determined to be monitored by the monitoring sub-time period It is awake.
5. 如权利要求1所述的方法,其特征在于,获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据之后,还包括: 对所述多个子监测时间段内的多轴加速度数据按照预设频率进行低通滤波,得到每个子监测时间段内的低频多轴加速度数据; 分别确定每个子监测时间段内低频多轴加速度数据的复杂度; 根据所述监测时间段内的所述多个子监测时间段内低频多轴加速度数据复杂度的均值和方差,确定所述监测时间段内该复杂度的动态阈值; 根据子监测时间段内低频多轴加速度数据的复杂度是否大于所述复杂度的动态阈值, 确定所述待监测者在该子监测时间段内为睡眠状态或清醒状态的第二睡眠分析结果; 针对第一睡眠分析结果为睡眠状态且第二睡眠分析结果为清醒状态的子监测时间段, 确定该子监测时间段的第三睡眠分析 After 5. The method according to claim 1, wherein the multi-axis acceleration data acquiring person to be monitored according to a preset sampling frequency acquired in each sub-period monitoring, further comprising: the plurality of sub-time monitoring multi-axis acceleration data in the segments according to a preset frequency of the low pass filter, the low frequency multi-axis acceleration data obtained for each sub-period monitoring; respectively determining the complexity of each sub-period monitoring low-frequency multi-axis acceleration data; according to the monitoring a plurality of sub-time period of the low frequency multi-axis acceleration monitoring period data mean and variance of the complexity of determining the threshold value of the monitored dynamic complexity time period; multi-axis acceleration data based on the low frequency sub-period monitoring complexity is greater than the complexity of a dynamic threshold to determine the person to be monitored in the sub-period monitoring sleep or awake sleep second analysis result; analysis results for the first sleep state and second sleep the results of two sleep awake monitoring sub period, the sub-sleep analysis to determine a third time period monitored 果为清醒状态。 Fruit is awake.
6. 如权利要求5所述的方法,其特征在于,确定一个子监测时间段内低频多轴加速度数据的复杂度,具体包括: 确定所述子监测时间段内低频多轴加速度数据的极值个数以及相邻的极大值与极小值之差; 对所述子监测时间段内低频多轴加速度数据的极值个数以及相邻的极大值与极小值之差,进行加权求和,确定子监测时间段内低频多轴加速度数据的复杂度。 6. The method according to claim 5, wherein determining the complexity of a monitoring period of the low frequency sub-multi-axis acceleration data comprises: determining the low frequency sub-period multi-axis acceleration monitoring data extremum the number and the difference between adjacent local maximum values ​​and the minimum values; the number of the extreme value of the low-frequency multi-axis acceleration monitoring sub-period data and the difference between adjacent local maximum values ​​and the minimum values, weighted summing, to determine the complexity of the monitoring period of the low frequency sub-multi-axis acceleration data.
7. 如权利要求5所述的方法,其特征在于,确定所述监测时间段内复杂度的动态阈值, 具体包括: 当所述监测时间段内复杂度的均值与方差之和大于第一预设复杂度阈值时,将所述第一预设复杂度阈值确定为所述监测时间段内复杂度的动态阈值; 当所述监测时间段内复杂度的均值与方差之和不大于第一预设复杂度阈值,且所述监测时间段内复杂度的均值与方差之和小于第二预设复杂度阈值时,将所述第二预设复杂度阈值确定为所述监测时间段内复杂度的动态阈值,其中,所述第二预设复杂度阈值小于所述第一预设复杂度阈值; 当所述监测时间段内复杂度的均值与方差之和不大于第一预设复杂度阈值,且所述监测时间段内复杂度的均值与方差之和不小于第二预设复杂度阈值时,将所述监测时间段内复杂度的均值和方差之和确定为所述监测时间段内 7. The method according to claim 5, wherein determining the complexity of the monitoring of the dynamic threshold period comprises: when the mean and variance of the complexity of the monitoring period is greater than a first pre- when the complexity of the set threshold, the first predetermined threshold value is determined as the complexity of the dynamic threshold value of the monitored period of time complexity; monitoring period when the complexity of the mean and variance is not greater than a first pre- when the complexity of the set threshold value, and the complexity of the monitoring period and the mean and variance of the complexity than a second preset threshold value, the second predetermined threshold value is determined as the complexity of the complexity of the monitoring period the dynamic threshold, wherein the second predetermined complexity threshold is smaller than the first predetermined complexity threshold; when the complexity of the monitoring period mean and variance is not greater than a first predetermined threshold complexity when the complexity of the monitoring and the mean and variance over a period of not less than a second predetermined complexity threshold, the complexity of the monitoring period and the mean and variance of said monitoring period is determined 杂度的动态阈值。 Dynamic threshold heteroaryl degrees.
8. —种睡眠分析装置,其特征在于,包括: 数据获取单元,用于获取按照预设采样频率在每个子监测时间段内采集的待监测者的多轴加速度数据,所述多轴加速度数据包括多个多轴加速度,其中,一个监测时间段包括多个子监测时间段; 活动量确定单元,用于分别基于在每个子监测时间段内采集的所述待监测者的多轴加速度数据,确定所述待监测者在每个子监测时间段内的活动量; 活动量特征值确定单元,用于分别针对每个子监测时间段,根据该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量,确定所述待检测者在每个子监测时间段内的活动直特征值,所述活动量特征值确定单元,具体用于确定该子监测时间段对应的时间段窗口包括的子监测时间段内的活动量的均值、方差,分别作为该子监测时间段对应的时间段窗口内活动量的 8. - kind of sleep analysis apparatus characterized by comprising: a data acquisition unit for acquiring data to be monitored by a multi-axis accelerometer in accordance with a predetermined sampling frequency acquired in each sub monitoring period, the multi-axis acceleration data comprising a plurality of multi-axis acceleration, wherein a monitoring period comprises monitoring a plurality of sub-time periods; activity determining unit configured to be monitored by a multi-axis acceleration data are acquired based on the monitoring in each sub-time period determined to be monitored by the amount of activity in each of the sub-period monitoring; activity characteristic value determining means for separately for each sub monitoring period, the monitoring period according to the sub-period corresponding to the sub-monitor window includes a time period of the amount of activity is determined to be detected activities monitored in each sub-period straight characteristic value, the characteristic value activity determination means for determining a particular subset of the monitoring sub-period period corresponding to the window comprises monitoring the amount of activity time period the mean, variance, respectively, within a monitoring period of the sub-time period corresponding to the amount of the active window 值、方差;确定该时间段窗口内活动量大于预设活动量的子监测时间段的个数;对该时间段窗口内活动量的均值、方差以及该时间段窗口内活动量大于预设活动量的子监测时间段的个数,进行加权求和,得到该子监测时间段内的活动量特征值;其中,子监测时间段对应的时间段窗口包括该子监测时间段以及其前后若干个子监测时间段; 动态阈值确定单元,用于根据监测时间段内所述多个子监测时间段内的活动量特征值,确定所述监测时间段内活动量特征值的动态阈值; 处理单元,用于分别将每个子监测时间段内的活动量特征值与所述动态阈值进行比较,得到所述待监测者在每个子监测时间段内为睡眠状态或清醒状态的第一睡眠分析结果。 Value, the variance; determining the period of time the active window is greater than the number of child monitoring a preset period of activity; the average amount of activity within the window period, and the variance of the active window is greater than the preset period Activity monitoring the number of sub periods amount, a weighted sum obtained amount of activity in the sub-period monitoring feature value; wherein, the monitoring sub-period period corresponding to the sub-window includes several sub monitored before and after the time period and monitoring period; dynamic threshold value determination unit for monitoring the amount of activity characteristic value according to the time period plurality of sub-monitoring periods, determining a dynamic threshold time period monitoring activity characteristic values; and a processing unit for respectively, each sub-period amount of activity monitoring feature value is compared with the dynamic threshold value, obtained by the analysis to be monitored first sleep or awake state to the sleep state in each sub-period monitoring results.
9.如权利要求8所述的装置,其特征在于,所述活动量确定单元,具体用于当该子监测时间段内采样点的多轴加速度大于预设加速度阈值时,确定所述待监测者在该采样点对应的时刻是活动的,所述采样点根据预设采样频率进行确定;以及将在该子监测时间段内确定所述待监测者活动的总次数,确定为所述待监测者在该子监测时间段的活动量。 9. The apparatus according to claim 8, wherein said activity determination unit is configured to, when the sub-period monitoring sampling points is greater than a predetermined acceleration threshold multi-axis acceleration value is determined to be monitored those corresponding to the sample point at a time is active, the sample point determined according to a predetermined sampling frequency; and determining the total number of the events to be monitored by the monitoring sub-time period, it is determined to be monitored by monitoring the amount of activity in the sub-periods.
10.如权利要求8所述的装置,其特征在于,所述动态阈值确定单元,具体用于: 确定监测时间段内所述多个子监测时间段内的多个活动量特征值的均值和方差,分别作为所述监测时间段内活动量特征值的均值和方差; 当所述监测时间段内活动量特征值的均值与方差之和大于第一预设活动量特征值阈值时,将所述第一预设特征值阈值确定为所述监测时间段内活动量特征值的动态阈值; 当所述监测时间段内活动量特征值的均值与方差之和不大于第一预设活动量特征值阈值,且所述监测时间段内活动量特征值的均值与方差之和小于第二预设活动量特征值阈值时,将所述第二预设活动量特征值阈值确定为所述监测时间段内活动量特征值的动态阈值,其中,所述第二预设活动量特征值阈值小于所述第一预设活动量特征值阈值; 当所述监测时间段内活动量特 10. The apparatus according to claim 8, wherein said dynamic threshold value determining unit is configured to: determining the mean and variance values ​​of characteristic amounts of a plurality of active periods of the plurality of local monitors monitoring period , respectively, mean and variance as the period of activity monitoring feature value; when the mean and variance of the monitoring feature value over a period of activity and an amount greater than the first predetermined threshold value characteristic when active, the wherein a first predetermined threshold value is determined as the dynamic threshold time period monitoring activity characteristic value; when the mean and variance of the monitoring feature value period of activity is not greater than a first predetermined value characteristic activity threshold, and the mean and variance of the characteristic value monitoring activity over a period of less than a second predetermined activity threshold value characteristic, the second predetermined threshold amount of activity is determined as the characteristic value of the monitoring period dynamic threshold value of the characteristic activity, wherein the second predetermined threshold value characteristic activity than the first predetermined threshold value characteristic activity; monitoring time period when the activity Laid 值的均值与方差之和不大于第一预设活动量特征值阈值,且所述监测时间段内活动量特征值的均值与方差之和不小于第二预设特征值阈值时,将所述监测时间段内活动量特征值的均值与方差之和确定为所述监测时间段内活动量特征值的动态阈值。 When the mean and variance values ​​and the sum is not greater than a first predetermined threshold value characteristic activity, and the activity monitoring feature value and the period mean and variance of the feature value is not less than a second predetermined threshold value, the activity monitoring period mean and variance of the feature value is determined as the sum of the monitoring period of the dynamic threshold value characteristic activity.
11. 如权利要求8所述的装置,其特征在于,所述处理单元,具体用于当子监测时间段的活动量特征值大于所述监测时间段内活动量特征值的动态阈值时,确定待监测者在该子监测时间段内为清醒状态。 11. The apparatus according to claim 8, wherein the processing unit, particularly for monitoring activity wherein when the sub-period dynamic threshold value is greater than the period of monitoring activity characteristic values, determined to be monitored by the monitoring period in the sub-awake state.
12. 如权利要求8所述的装置,其特征在于,所述装置,还包括:低频数据处理单元,用于对所述多个子监测时间段内的多轴加速度数据按照预设频率进行低通滤波,得到每个子监测时间段内的低频多轴加速度数据; 分别确定每个子监测时间段内低频多轴加速度数据的复杂度; 根据所述监测时间段内的所述多个子监测时间段内低频多轴加速度数据复杂度的均值和方差,确定所述监测时间段内该复杂度的动态阈值; 根据子监测时间段内低频多轴加速度数据的复杂度是否大于所述复杂度的动态阈值, 确定所述待监测者在该子监测时间段内为睡眠状态或清醒状态的第二睡眠分析结果; 针对第一睡眠分析结果为睡眠状态且第二睡眠分析结果为清醒状态的子监测时间段, 确定该子监测时间段的第三睡眠分析结果为清醒状态。 12. The apparatus according to claim 8, characterized in that said apparatus further comprising: a low frequency data processing unit for monitoring multi-axis acceleration data of the plurality of sub-time period according to a preset low-pass frequency filtering, to obtain a low-frequency multi-axis acceleration monitoring data in each sub-period, respectively; determining the complexity of each sub-period monitoring data of low frequency multi-axis acceleration; monitoring period based on the plurality of low frequency sub-period of the monitoring Multi-axis acceleration data mean and variance of the complexity of determining the dynamic threshold time period complexity of the monitoring; is greater than a dynamic threshold based on the complexity of the complexity of the monitoring period of the low frequency sub-multi-axis acceleration data is determined to be monitored by said monitoring period in the sub-awake state to the sleep state or the second sleep analysis; for the first sleep analysis result to the sleep state and the second sub-monitor sleep analysis result is awake period, determined the results of the third sub-sleep monitoring period was awake.
13. 如权利要求12所述的装置,其特征在于,所述低频数据处理单元,确定一个子监测时间段内低频多轴加速度数据的复杂度,具体用于确定所述子监测时间段内低频多轴加速度数据的极值个数以及相邻的极大值与极小值之差;以及对所述子监测时间段内低频多轴加速度数据的极值个数以及相邻的极大值与极小值之差,进行加权求和,确定子监测时间段内低频多轴加速度数据的复杂度。 13. The apparatus of claim 12, wherein said low frequency data processing unit determines the complexity of a monitoring period of the low frequency sub-multi-axis acceleration data, the specific monitoring time period for determining the low-frequency sub the number of extrema multi-axis acceleration data and the difference between adjacent local maximum values ​​and the minimum values; and monitoring extreme value of the sub-period the number of low frequency multi-axis acceleration data and the maximum value adjacent to the the difference between the minimum value, the weighted sum to determine the complexity of the monitoring period of the low frequency sub-multi-axis acceleration data.
14. 如权利要求12所述的装置,其特征在于,所述低频数据处理单元,确定所述监测时间段内复杂度的动态阈值,具体用于当所述监测时间段内复杂度的均值与方差之和大于第一预设复杂度阈值时,将所述第一预设复杂度阈值确定为所述监测时间段内复杂度的动态阈值; _ ^ 当所述监测时间段内复杂度的均值与方差之和不大于第一预设复杂度阈值,且所述监测时间段内复杂度的均值与方差之和小于第二预设复杂度阈值时,将所述第二预设复杂度阈值确定为所述监测时间段内复杂度的动态阈值,其中,所述第二预设复杂度阐值小于所述第一预设复杂度阈值; 当所述监测时间段内复杂度的均值与方差之和不大于第一预设复杂度闕测时间段内复杂度的均值与方差之和不小于第二预设复杂度阈值时,将所述时间段内复杂度的均值和方差之和确定为所述监测时 14. The apparatus of claim 12, wherein said low frequency data processing unit, determining a dynamic threshold complexity of the monitoring period, when the mean value for the specific monitoring period and complexity when the variance is greater than a first predetermined threshold complexity, the complexity of the first predetermined threshold value as the monitored dynamic threshold period of time complexity; _ ^ when the mean complexity of the monitoring period when the variance is not greater than a first predetermined threshold complexity, and complexity of the monitoring period and the mean and variance of the complexity than a second preset threshold value, the second predetermined complexity threshold determination monitoring said dynamic threshold complexity period, wherein the second predetermined value is less than the complexity explain the first predetermined complexity threshold; when the mean and variance of the complexity of the monitoring period and when no greater than a first predetermined period of time complexity Que measured mean and variance of the complexity and complexity is not less than a second predetermined threshold value, the time period complexity of the mean and variance is determined as the sum of the when said monitoring 段内复杂度的动态阈值。 The dynamic threshold segment complexity.
CN201310687525.2A 2013-12-16 2013-12-16 An analysis method and device sleep CN104706318B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310687525.2A CN104706318B (en) 2013-12-16 2013-12-16 An analysis method and device sleep

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310687525.2A CN104706318B (en) 2013-12-16 2013-12-16 An analysis method and device sleep

Publications (2)

Publication Number Publication Date
CN104706318A CN104706318A (en) 2015-06-17
CN104706318B true CN104706318B (en) 2018-02-23

Family

ID=53406402

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310687525.2A CN104706318B (en) 2013-12-16 2013-12-16 An analysis method and device sleep

Country Status (1)

Country Link
CN (1) CN104706318B (en)

Families Citing this family (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106580332A (en) * 2015-10-20 2017-04-26 中国移动通信集团公司 Motion detection method, device and equipment
CN105559751B (en) * 2015-12-14 2018-06-12 安徽华米信息科技有限公司 Micro-active state monitoring method, device and wearable device
CN105496362B (en) * 2015-12-15 2018-08-31 成都麦杰康科技有限公司 Sleep monitoring system and method
CN106913313A (en) * 2015-12-28 2017-07-04 中国移动通信集团公司 Sleep monitoring method and system
CN105640508B (en) * 2016-03-30 2018-09-18 安徽华米信息科技有限公司 Real-time sleep monitoring method and device, a smart wearable devices
CN106175696B (en) * 2016-09-14 2019-02-01 广州视源电子科技股份有限公司 Sleep state monitoring method and system
CN106388786A (en) * 2016-12-14 2017-02-15 广东技术师范学院 Processing method acquiring data based on intelligent ring
TWI642409B (en) * 2017-05-22 2018-12-01 研鼎智能股份有限公司
CN108606882A (en) * 2018-03-23 2018-10-02 合肥工业大学 Intelligent wheelchair control system based on myoelectricity and acceleration self-adaptation control

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6527715B2 (en) * 1998-10-30 2003-03-04 The United States Of America As Represented By The Secretary Of The Army System and method for predicting human cognitive performance using data from an actigraph
US7189204B2 (en) * 2002-12-04 2007-03-13 Cardiac Pacemakers, Inc. Sleep detection using an adjustable threshold
US7996076B2 (en) * 2007-04-02 2011-08-09 The Regents Of The University Of Michigan Automated polysomnographic assessment for rapid eye movement sleep behavior disorder
US20100100004A1 (en) * 2008-10-16 2010-04-22 Koninklijke Nederlandse Akademie Van Wetenschappen Skin Temperature Measurement in Monitoring and Control of Sleep and Alertness
US8280501B2 (en) * 2010-04-16 2012-10-02 Dyna Dx Corporation Systems and methods for quantitatively characterizing slow wave activities and states in sleep
JP5760669B2 (en) * 2011-05-12 2015-08-12 富士通株式会社 State determining apparatus, and a state determination method

Also Published As

Publication number Publication date
CN104706318A (en) 2015-06-17

Similar Documents

Publication Publication Date Title
Bravi et al. Review and classification of variability analysis techniques with clinical applications
Wilson et al. Spike detection: a review and comparison of algorithms
Adeli et al. A wavelet-chaos methodology for analysis of EEGs and EEG subbands to detect seizure and epilepsy
Cole et al. Intrinsic and task-evoked network architectures of the human brain
Zalesky et al. Towards a statistical test for functional connectivity dynamics
US20070060831A1 (en) Method and system for detecting and classifyng the mental state of a subject
US20080234785A1 (en) Sleep controlling apparatus and method, and computer program product thereof
US20090062679A1 (en) Categorizing perceptual stimuli by detecting subconcious responses
Parra et al. Response error correction-a demonstration of improved human-machine performance using real-time EEG monitoring
Breakspear et al. A novel method for the topographic analysis of neural activity reveals formation and dissolution of ‘dynamic cell assemblies’
Motamedi-Fakhr et al. Signal processing techniques applied to human sleep EEG signals—A review
Valenza et al. Dominant Lyapunov exponent and approximate entropy in heart rate variability during emotional visual elicitation
US8400313B2 (en) Vehicle driver sleep state classification generating device based on Hidden Markov Model, sleep state classification device and warning device
CN101310242A (en) Method and system for detecting and classifying facial muscle movements
WO2009024273A1 (en) Method and system for monitoring sleep
Chen et al. Automatic detection of alertness/drowsiness from physiological signals using wavelet-based nonlinear features and machine learning
JP4680252B2 (en) How to sleep evaluation device and sleep evaluation
WO2008065724A1 (en) Arousal level judging method and arousal level judging program
JP2010068848A (en) Drowsiness determination apparatus and program
US20150190086A1 (en) Automated sleep staging using wearable sensors
Nallathambi et al. Integrate and fire pulse train automaton for QRS detection
US20180020953A1 (en) Method and device for identifying human movement state
Myers et al. Testing sensory evidence against mnemonic templates
CN103815896A (en) Mental fatigue monitoring method, device and system and mobile processing terminal
Ren et al. Off-line and on-line stress detection through processing of the pupil diameter signal

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
GR01