//Logo Image
Author: Chih-Ming Cheng, Yeh-Liang Hsu, Chang- Ming Young (2008-02-21); recommended: Yeh-Liang Hsu (2010-06-12).
Note: This paper is published in Telemedicine and e-Health, Vol. 14, No. 10, pp. 1044-1056, December 2008.

Development of a portable device for tele-monitoring of physical activities during sleep

Abstract

Low motor activity levels and prolonged episodes of uninterrupted immobility are characteristics of sleep. In clinical practice, the use of polysomnographic (PSG) recording is a standard procedure to assess sleep. However, PSG is not suitable for long term monitoring in the home environment. This paper describes the development of a portable tele-monitoring device that detects movements of a subject by conductive mats, and evaluates sleep stages via physical activity data. The device itself also serves as a web server. Doctors and caregivers can access real-time and historical data via an IE browser or a remote application program for tele-monitoring of physical activities and sleep/awake states during sleep, while the patients stay in their own homes.

In our validation test with 4 normal subjects and 4 arousal subjects, this system showed a good performance in locating sleep epochs of a subject. The sensitivity of locating sleep epochs was 89.5% and the average positive prediction value was 94.8%, with a specificity of 84.3%. This device is not intended to be a diagnosis device, instead, it is to be used as a home telehealth tool for monitoring physical activity and sleep/awake states. This portable tele-monitoring device provides a convenient approach to better understand and recognize a subject’s sleep pattern through long-term sleep monitoring in the home environment.

Keywords: home telehealth, physical activity, sleep pattern.

1.     Introduction

During sleep, low motor activity levels and prolonged episodes of uninterrupted immobility are associated with increasing sleep depth. There are also motor disturbances that are triggered by sleep such as restless legs syndrome (RLS) and periodic limb movements during sleep (PLMS). Such symptoms disrupt sleep and cause daytime tiredness and sleepiness.

Several non-invasive and unrestrained sensing techniques have been developed for the monitoring of body movement during sleep. The use of load cells or force sensors is the most common approach to detect body movements in bed. Load cells represent a simple and durable technology, which was used by several researchers [1-4]. The use of force sensor is also a popular technique for monitoring body movements in bed. Nishida et al. [5] presented the idea of a robotic bed, which is equipped with 221 pressure sensors for monitoring of respiration and body position. Van der Loos et al. [6] also proposed a similar system called SleepSmart™, composed of a mattress pad with 54 force sensitive resistors and 54 resistive temperature devices, to estimate body center of mass and index of restlessness. These large-size equipments are not easy to set up and can be used in specific laboratories only.

Many pad-based solutions have been proposed. Several authors have employed the static charge sensitive bed (SCSB) for monitoring of motor activity. The SCSB is composed of two metal plates with a wooden plate in the middle that must be placed under a special foam plastic mattress, which operates like a capacitor [7-9]. Watanabe et al. [10] designed a pneumatics-based system for sleep monitoring. A thin, air-sealed cushion is placed under the bed mattress of the subject and the small movements attributable to human automatic vital functions are measured as changes in pressure using a pressure sensor.

Other sensing techniques, such as optical fibers and conductive fibers have also been used for monitoring body movement in bed. Tamura [11] et al. proposed a body movement monitoring system using optical fibers. Kimura et al. [12] designed an unobtrusive vital signs detection system, which uses conductive fiber sensors to detect body position, respiration, and heart rate. These fiber sensors can be incorporated in a conventional bed sheet for home use.

Polysomnography (PSG) is considered to be the “gold standard” method for assessing sleep. Different sleep stages are evaluated by medical specialists using PSG data such as electroencephalogram (EEG), electrooculogram (EOG), and electromyogram (EMG) based on the Rechtschaffen and Kales (R-K) method. This technique usually requires individuals to sleep in research laboratories which are known to change habitual sleep patterns [13] and induce a “first-night” effect [14].

Since the development of the actioculographic monitor in 1979, a novel method to estimate sleep stage through body movements have been suggested. Measurement of motility has become a popular method in the study of human sleep. Kayed et al. [15] used three criteria, eye movements, body movements, and submental electromyogram, to identify wake, REM sleep, and NREM sleep. Based on this development, a wearable wrist actigraphy, has been developed for the identification of wake, REM and NREM stages [16-18]. Ajilore et al. [19] proposed a home-based sleep monitoring system, called Nightcap, which uses eyelid and body movement sensors to discriminate awake, NREM, and REM sleep automatically. Literature shows that actigraphy is a valuable device to detect sleep-wake period. The agreement between actigraphy and PSG with epoch-to-epoch comparison, ranged from 80% to 90% [16, 20]. Although estimating sleep stages through body movement is not as accurate as PSG, studies are in general agreement that measures with body movement were fairly sensitive in detecting sleep.

Similar to wrist actigraphy, Choi et al. [21] introduces a bed actigraphy for user-friendly sleep-wake monitoring. An automatic scoring algorithm scores each epoch of the recordings for either ‘wake’ or ‘sleep’ by the signals of 4 load cells, which are installed at each corner of the bed. Bed actigraphy has some advantages over ordinary actigraphic devices. First, the bed actigraphy does not constrain the subjects during recording. The subject can even be unaware of the recording process. Secondly, bed actigraphy can detect the overall movements of the subject by measuring the body activity at four points, while an actigraph detects only the movements of the body part on which it is mounted. The bed actigraphy device developed by Choi et al. showed good agreement (95.2%) but medium sensitivity and positive predictive value (64.4%, 66.8%).

For the needs of remote monitoring of sleep, several research activities are underway. Kristo et al. [22] presented a telemedicine protocol for the online transfer of PSGs from a remote site to a centralized sleep laboratory, which provided a cost-saving approach for the diagnosis of OSAS. Seo et al. [23] developed a non-intrusive health-monitoring house system to monitor patients’ electrocardiogram results, weight, movement pattern and snoring, in order to get the information of patients’ health status and sleep problems. Choi et al. [24] presented a ubiquitous health monitoring system in a bedroom, which monitors ECG, body movements and snoring with non-conscious sensors.

A centralized framework is used in most home telehealth systems, in which a centralized database is used for data storage and analysis. Figure 1 shows the structure of the decentralized home telehealth system developed by the authors [25]. Instead of using a centralized database that gathers data from many households, a single household is the fundamental unit for sensing, data transmission, storage and analysis. The core of the system is the “Distributed Data Server (DDS)” inside a household, which is a thin server designed specifically for the decentralized home telehealth system. It consists of a PIC server mounted on a peripheral application board. The PIC server integrates a PIC microcontroller (PIC18F6722, Microchip), EEPROM (24LC1025, Microchip) and a networking IC (RTL8019AS, Realtek). The DDS provides networking capability and can be used as a web server.

As shown in Figure 1, sensing data from sensors embedded in the home environment are transmitted to the DDS, and are then processed and stored in the Multi-Media-Card (MMC) of the DDS. Authorized remote users can request data from the DDS using an Internet web browser (running Java applets) or a Visual Basic (VB) program. Event-driven messages (mobile phone text messages or emails) can be sent to specified caregivers when an urgent situation is detected.

Figure 1. The structure of the decentralized home telehealth system

There are several advantages of the decentralized structure over the traditional centralized database structure:

(1)    The scale of the home telehealth system is much smaller, which makes it economically viable and acceptable to the end-users.

(2)    Instead of sending the health monitoring data to a centralized database in a home healthcare provider, health monitoring data are stored within the household. Only authorized caregivers can access the data. Privacy is better protected.

(3)    The route from the sensor to server is much shorter. Data transmission is easier and more reliable. When the Internet communication fails, the local system can still function normally and keep collecting data. Thus data integrity is better preserved.

(4)    The home telehealth provider only needs to maintain an application server to provide email, short message, and Java services. The DDS can also be used as a gateway if a centralized database is needed.

This paper describes the development of a “Physical Activity Detecting Mat (PAD-Mat)” based on the decentralized home telehealth system structure described above. Figure 2 shows the structure of the PAD-Mat developed in this research. Similar to the decentralized structure in Figure 1, the core component of the PAD-Mat is a DDS that detects activity signals from upper limb, body and leg, captured by three conductive mats. Doctors and caregivers can access the DDS for real-time and historical data via an IE browser or a remote application program for tele-monitoring of physical activities and sleep/awake states during sleep, while the patients stay in their own homes. If the subject in bed has no activities for a predefined period of time, an event message (mobile phone short messages or emails) can be sent to specified caregivers.

Figure 2. Structure of the PAD-Mat

The PAD-Mat provides a non-constrained and non-conscious approach for tele-monitoring of physical activities during sleep. It can be used for in-bed detection and monitoring motor disturbances, such as RLS and PLMS. In addition, using the sleep activity index (SAI) proposed in this research, sleep and wakeful stages can be classified on-line.

The rest of this paper is organized as follows. Section 2~4 present the design and network framework of the PAD-Mat. Section 5 discusses the validation results of the PAD-Mat. In summary, Section 6 describes the applications and potential benefit of the PAD-Mat.

2.     Design of the PAD-Mat

As shown in Figure 2, three conductive mats placed under the chest, hip, and legs are used in the PAD-Mat to detect the physical activities with the resistance changes of the mats. The signals are analyzed with the physical activity detecting algorithms. The following sections describe the design of the conductive mats, the in-bed detecting algorithm, the physical activity evaluating algorithm and the sleep/awake identifying algorithm in details.

2.1  Design of the conductive mats

The conductive mats are made of conductive fabric [Ming Young Biomedical Corp., Taiwan]. There are two types of conductive mat as shown in Figure 3 and 4. The PAD-Mat is composed of two Type-A mats for detecting upper limb and leg activities (the “upper limb mat” and the “leg mat”) and one Type-B mat for detecting body activities (the “body mat”). Physical activities are captured as the electrical resistance changes with motion on each mat. The construction of the conductive mat is inspired by the intestinal villi, which increase the surface area of absorption. Sensitivity of this bended structure is significantly higher than that of a straight structure. A layer of non-conducting foam is inserted in the Type-B mats so that the electrical resistance of the Type-B conductive mats can sufficiently restore after the subject leaves the bed. A simple dividing circuit is used to converts resistance changes of the body mat into voltage changes. Figure 5 shows the placement of the PAD-Mat on the bed.

Figure 3(a). Appearance of conductive mats (Type-A)

Figure 3(b). Dimensions of conductive mats (Type-A)

Figure 4(a). Appearance of conductive mats (Type-B)

Figure 4(b). Appearance of conductive mats (Type-B)

Figure 5. Placement of the PAD-Mat

The analog signals are digitized at a sampling frequency of 2KHz with a 10-bit A/D converter in the PIC server. Figure 6(a) shows a 16-min record of the body mat input signals. The subject changed body position at around the 2nd minute, 8th minute, and 13th minute. A series of smoothing and simplifying procedures further processes these signals. A moving average filter, with a window size of 20, smoothes the profile of the input voltage values. After that, the signals are further simplified into 10 data points per second to save computational resource in the following detecting algorithms (Figure 6(b)). This procedure is applied to signals from all 3 conductive mats.

Figure 6(a). Input voltages of body activities

Figure 6(b). Smoothed and simplified values of the input voltage

2.2  In-bed detecting algorithm

The body mat is also used to detect whether the subject is in bed or not. The input voltage decreased when the subject lies on the mat, and stays at 3.3 V after the subject leaves the mat. With a threshold limit 3.2V, a data is coded “1” if it is below the threshold limit and a “0” if the slope value is above the threshold limit. Figure 7 shows a one-night record of the input voltage of the body mat. The subject went to bed at around 2:35 AM, and left bed at around 8:15 AM. Body movements might cause transient increases of voltage which results in a rapid switch from “0” to “1”. To avoid that problem, the PAD-Mat uses one minute as the time unit to define the “In-bed Code”. If there are more “1” than “0” in that minute, the In-bed Code of the minute is coded “1” and vice versa. As shown in Figure 7, the PAD-Mat recognizes the input voltage values as a series of In-bed Codes and the total in-bed time is 340 minutes.

Figure 7. In-bed detection by the body mat

2.3  Physical activity detecting algorithm

After the smoothing and simplifying procedures, the physical activity detecting algorithm analyzes the data and output activity indexes by the following steps:

(1)   Calculate the absolute slope values of activity signals in Figure 6(b), as shown in Figure 8(a).

(2)   Symbolize a slope value with an activity code “1” if it is above the threshold limit and a “0” if the slope value is under the threshold limit. Figure 8(b) shows a 10-second clip of Figure 8(a), a body movement was recognized as a series of Activity Codes “1”, using a threshold limit 0.025. The threshold limit for upper limb and leg activities are both 0.02.

(3)   Tally the counts of “1” series each minute as the “Body Activity Index (BAI)”, which represents the length of activities occurred in the past minute. Figure 8(c) shows the body activity index versus time of Figure 6(a). 3 body movements, lasted 1.5, 1.2, and 1.5 seconds, have been recognized.

Figure 8(a). Absolute slope values of body movement of a subject (16 minutes)

Figure 8(b). Activity code of a body movement (Threshold limit value=0.02)

Figure 8(c). The body activity index versus time

Following the same procedures, the PAD-Mat also tallies the “Upper Limb Activity Index (ULAI)” and the “Leg Activity Index (LAI)” for each minute. Figure 9 shows a 7.5-hour record of the ULAI, BAI and LAI of a subject evaluated by PAD-Mat. ULAI, LAI and BAI, which represent the duration of activities happened in a minute, are quantitative indexes for assessing physical activities. These indexes can be used for the diagnosis of motor disturbances that are triggered by sleep such as RLS and PLMS.

Figure 9. Activity indexes of a subject from 0:30 to 8:00 AM

3.     The sleep/awake distinguishing algorithm

In sleep, motor activity is reduced in comparison to the waking state. The frequency of all movements decreases with depth of sleep, with progressive decrease in the number of movements from sleep stage I to stage IV [26].

Since the late 1970s, a growing number of studies have demonstrated the validity of actigraphy in distinguishing between sleep and wakefulness. Jean-Louis et al. [27] utilizes a simple technique for scoring sleep/awake epochs. The epochs containing higher activity than a given threshold were scored as wake and lower activity as sleep, and arousals lasting 3 minutes or less were rescored as sleep. Cole et al. [28] proposed two rules for identifying sleep and awake epochs: (1) The sleep of 1, 3, or 4 minutes was rescored as wake if it preceded at least 4, 10, or 15 minutes of wake, respectively; (2) The sleep of 6 or 10 minutes surrounded by at least 10 or 20 minutes of wake was rescored as wake, respectively.

Sadeh et al. [29] used logistic regression analysis for the variables of body activity while the sleep/awake classification of PSG acted as the dependent variable in the analysis. Five activity variables were computed for each epoch in the activity signals: original value, mean, standard deviation, number of epochs above a specified activity level, and the natural logarithm. Most actigraphy related devices scored signals using similar algorithms to classify sleep/awake epochs, based on variables of body activity.

Rachwalski et al. [30] used a pressure pad placed below a person’s hips to measure activity in bed. Activity in bed is measured by changes in the pressure measurements. The pressure measurements aggregated into 30-second epochs by averaging the data every 30 seconds. Awakenings (or periods of restlessness) are defined by 3 consecutive minutes of body movements. Choi et al. [21] designed a bed actigraphy system for distinguishing between sleep and awake. Signals were coded as “1” as the intensity of signals higher than the threshold. If the duration of “1” is longer than 3 seconds in an epoch (30 second length), the epoch is scored as “Wake”.

To sum up, both the quantity and tendency of physical activity are considered as the character of physical activity and used for the evaluating of level of wakefulness. To identify the subject’s sleep condition in real time, ULAI, BAI and LAI described in the previous section are used to recognize whether the subject is asleep in the PAD-Mat. To consider the tendency of activity, weighted activity indexes are calculated by considering the activity indexes of the past minutes with different weighting, as shown in Equation (1). For example, weighted-BAI of the 5th minute is calculated as Equation (2). Weighted-ULAI and weighted-LAI are also calculated with the same equation.

                                          (1)

          (2)

To explain the relationship between physical activities and sleep depth, sleep reports of PSG and physical activity data of the PAD-Mat were collected from 5 subjects. A “Sleep Value (SV)” was defined based on the PSG report and scored by minute. The SV of the minute is coded “0” if PSG recognized 2 epochs (1 minute) as awake, and “1” as 2 sleep epochs were detected. If there were only 1 sleep epoch in the past minute, the Sleep Value was coded “0.5”. Regression analyses were used to explain the relationship between physical activities and SV. Regression analyses were used on three 80-minute epochs randomly selected from each subject, with weighted activity indexes and different k values in Equation (1). For example, k=5 means the regression analysis model includes weighted-ULAI, weighted-BAI and weighted-LAI, which were calculated using the raw data of the past 5 minutes.

Table 1 shows that result of regression analyses performed with different k values. ULAI was found to be non-significant with different k values, which means ULAI is not a valuable for sleep/awake classifying. Furthermore, BAI and LAI are both highly significant with the k value of 5. Equation (3) is the regression equation used in this research, in which BAI and LAI were considered for evaluating sleep with the k value of 5.

    SV = 0.6590.028 W-BAI0.026 W-LAI                                             (3)

Figure 10 shows an 8 hour recording of W-BAI, W-LAI and Sleep Value calculated with Equation (1), (2) and (3).

Table 1. Regression analyses with different k values

k value

Significant

W-ULAI

W-BAI

W-LAI

1

N (P=0.919)

N (P=0.195)

N (P=0.214)

2

N (P=0.125)

N (P=0.559)

Y (P=0.032)

3

N (P=0.082)

N (P=0.085)

Y (P=0.079)

4

N (P=0.087)

Y (P=0.010)

N (P=0.258)

5

N (P=0.390)

Y (P=0.008)

Y (P=0.002)

6

N (P=0.165)

Y (P=0.049)

Y (P=0.019)

Figure 10. W-BAI, W-LAI and Sleep Value of a subject from 0:00 to 8:00AM

Figure 11 shows an 8-hour recording (0:00 to 8:00 AM) of In-Bed Codes and Sleep Codes of the same data in Figure 10, as well as the final output “Sleep State”. In this research, the Sleep Code of the past minute is assigned as “1” if SV of this minute calculated by Equation (3) is higher than 0.5, and “0” if SV is under the threshold limit 0.5.

As shown in Figure 11, “Sleep States” were coded “2” (sleep) if both Sleep Code and In-bed Code are “1”, and “0” (Empty Bed) if both Sleep Code and In-Bed Code are “0”. Sleep State1” (Awake) is recognized as the subject is in bed but stayed awake (In-bed Code 1 and Sleep Code 0). According to the output of the PAD-Mat in Figure 11, this subject went to bed at about 0:34 AM and fell asleep at 0:40 AM. After a 6-hour sleep, the subject woke up at about 6:50 AM and left bed at 6:50 AM.

Figure 11. In-bed Code, Sleep Code and Sleep States evaluated by the PAD-Mat

4.     User Interface and Network Framework

Figure 12 shows the network framework for the PAD-Mat. Doctors and caregivers can access real-time and historical data via a remote VB program. Figure 13 shows the VB interface window designed for the PAD-Mat. After typing an IP address and clicking the “real-time” button, real-time information of physical activities per minute and sleep state of the subject are displayed on the left side of the interface window. Three sleep states, sleep, awake and empty bed will be identified.

Doctors and caregivers can monitor 3 patients on the same window. Selecting one of the IP addresses on the left side, the user can also download historical data from the PAD-Mat at this IP address (select a date and time interval on the right side of the interface window and click the “download” button). Diagnosis reports for time to lie on bed and fall asleep, as well as total in-bed time and sleep duration within the selected time interval can be tallied automatically. Charts of activity indexes of upper limb, body and leg per minute versus time, and sleep states versus time can also be plotted for further analysis.

Besides motor disturbances monitoring and sleep states assessment, the PAD-Mat can be used for detecting body position changes. Bed sores are common problems in nursing homes. Usually caregivers have to change the subject’s body position every 2 hours. Figure 14 shows a 3-day history of a subject in a nursing home. Caregivers and family members can access the PAD-Mat from the Internet for tracking body activities of the patients.

Figure 12. A framework for the PAD-Mat

Figure 13. VB interface window for the Pad-Mat

Figure 14. Three-day history of a subject in a nursing home

5.     Validation test of the PAD-Mat

A validation test was designed to evaluate the performance of PAD-Mat. The PAD-Mat was used to detect sleep epochs of 4 normal sleepers and 4 sleepers with arousal in a full-night’s sleep. Sleep states judged by the PAD-Mat with Sleep Codes and In-bed Codes, were compared epoch-to-epoch with the diagnosis reports by a PSG. Figure 15 and Figure 16 show the sleep states scored via PAD-Mat and PSG of a subject in non-arousal group B and a subject in arousal group A. The subject of non-arousal group B did not leave bed until 7 AM, and the subject of arousal group A went to the bathroom at around 5 AM for about 4 minutes.

Figure 15. Sleep States scored via PAD-Mat and PSG (Non-arousal group B)

Figure 16. Sleep States scored via PAD-Mat and PSG (Arousal group A)

Table 2 shows the comparison results of locating sleep epochs using the Pad-Mat and PSG. A one-minute epoch was coded True Positive (TP) if the PAD-Mat and PSG both classified as sleep epoch, and True Negative (TN) if the PAD-Mat and PSG both classified as non-sleep epoch (awake and empty bed). Similarly, a segment was coded False Positive (FP) or False Negative (FN) if only the PAD-Mat or only the PSG recognized the state as sleep. Furthermore, in Table 2, sensitivity is defined as TP/(TP+FN), the proportion of sleep epochs correctly identified by the Pad-Mat and the total sleep epochs identified by PSG. The positive predictive value (PPV) is defined as TP/(TP+FP), which refers to the proportion of sleep epochs correctly identified by the Pad-Mat and the total sleep epochs identified by the PatMat. The specificity is a statistical measure of how well a binary classification test correctly identifies the negative cases, and is defined as TN/(FP+TN). In our validation test, the sensitivity of locating sleep epochs was 89.5% and the average PPV was 94.8%, with a specificity of 84.3%.

Table 2. System performance validation results of locating sleep segments

Group

ID

Sleep Time

TP

FN

FP

Sensitivity

PPV

Non-Sleep Time

TN

Specificity

Normal

A

372

261

28

8

90.3%

97.0%

83

75

90.4%

B

438

275

42

20

86.8%

93.2%

121

101

83.5%

C

402

296

25

2

92.2%

99.3%

81

79

97.5%

D

554

320

40

39

88.9%

89.1%

194

155

79.9%

Arousal

A

622

426

68

20

86.2%

95.5%

128

108

84.4%

B

399

283

8

32

97.3%

89.8%

108

76

70.4%

C

479

319

68

13

82.4%

96.1%

92

79

85.9%

D

267

259

8

0

97.0%

100%

70

66

94.3%

8 subjects

3533

2439

287

134

89.5%

94.8%

877

739

84.3%

6.     Conclusions

Since the invention of actigraphy, a growing number of studies have demonstrated how physical activities during sleep can be used in distinguishing between sleep and wakefulness. This is also a popular method for designing sleep monitoring devices used in the home environment.

In this research, the PAD-Mat, a portable device for tele-monitoring of physical activities was developed to evaluate body movements with quantitative measurement and recognize “sleep”, “awake” and “empty bed” states. In our validation tests, the Pad-Mat shows good sensitivity and specificity in identifying sleep states, and can be a non-constrain approach to better understand the sleep pattern.

The PAD-Mat system is designed for long-term monitoring at home based on the decentralized home telehealth system structure. What sets the structure apart from most other systems is the focus on a highly decentralized monitoring model and the portable nature of the system. We believe that this is the approach that is needed to make such systems economically viable and acceptable to the end-users.

Reference

Wheatley, D.W., Berme, N., and Ferguson-Pell, M.W., 1980, “A low-profile load transducer for monitoring movement”, Experimental Mechanics, v. 20, n. 6, pp.19-20.

Barbenel, J.C., Ferguson-Pell, M.W., and Beale, A.Q., 1985, “Monitoring the mobility of patients in bed”, Medical and Biological Engineering and Computing, v. 23, n. 5, pp. 466-468.

Choi, J.M.; Choi, B.H.; Seo, J.W.; Sohn, R.H.; Ryu, M.S.; Yi, W.; Park, K.S., 2004, “A System for Ubiquitous Health Monitoring in the Bedroom via a Bluetooth Network and Wireless LAN”, Engineering in Medicine and Biology Society, 2004. EMBC 2004. Conference Proceedings. 26th Annual International Conference of the IEEE, v. 2, pp. 3362-3365.

Adami, A.M., Hayes, T.L., Pavel, M., Singer, C.M., 2005, “Detection and classification of movements in bed using load cells”, IEEE 27th Annual International Conference of the Engineering in Medicine and Biology Society, pp.589-592.

Nishida, Y., Takeda, M., Mori, T., Mizoguchi, H. and Sato, T., 1997, “Monitoring patient respiration and posture using human symbiosis system,” Proceedings of the 1997 IEEE International Conference on Intelligent Robots and Systems, v. 2, pp. 632-639.

Van der Loos, H.F., Ullrich, N., and Kobayashi, H., 2004, “Development of Sensate and Robotic Bed Technologies for Vital Signs Monitoring and Sleep Quality Improvement”, Autonomous Robots, v. 15, n. 1, pp.67-79.

Alihanka, J., and Vaahtoranta, K., 1979, “A Static Charge Sensitive Bed - A New Method for Recording Body Movements during Sleep”, Electroencephalography and Clinical Neurophysiology, v. 46, pp. 731-734.

Rauhala, E., Erkinjuntti, M., and Polo, O., 1996, “Detection of Periodic Leg Movements with a Static-Charge-Sensitive Bed”, Journal of Sleep Research, v. 5, pp. 246-250.

Kaartinen, J., Kuhlman, I., and Peura, P., 2003, “Long-term Monitoring of Movements in Bed and Their Relation to Subjective Sleep Quality”, Sleep and Hypnosis, v. 5, n. 3, pp. 145-153.

Watanabe, K., Watanabe, T., Watanabe, H., Ando, H., Ishikawa, T., and Kobayashi, K., 2005, “Noninvasive measurement of heartbeat, respiration, snoring and body movements of a subject in bed via a pneumatic method”, IEEE Transactions on Biomedical Engineering, v. 52, pp. 2100-2107.

Tamura, T., Nishigaichi, A., and Nomura, T., 1992, “Monitoring of body movement during sleep in bed,” Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, v.14, pp. 1483-1484.

Kimura, H., Kobayashi, H., Kawabata, K., and Van der Loos, H.F., 2004, “Development of an unobtrusive vital signs detection system using conductive fiber sensors”, Proceedings. 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems, v. 1, pp. 307- 312.

Reynolds, C. F., III Grochocinski, V. J., Monk, T. H., Buysse, D. J., Giles, D. E., Coble, P. A., Matzzie, J.V., Doman, J., Monahan, J., and Kupfer, D.J., 1992, “Concordance between habitual sleep times and laboratory recording schedules”, Sleep, v. 15, pp. 571–575.

Riedel, B. W., Winfield, C. and Lichstein, K. L., 1998, “Anxiety and first night effect in older insomniacs”, Sleep, v.21, pp. 131.

Kayed, K., Hesla, P.E., and Rosjo, O., 1979, “The actioculographic monitor of sleep”, Sleep, v. 2, n. 2, pp.253-260.

Hauri, P.J.and Wisbey. J., 1992, “Wrist actigraphy in insomnia”, Sleep, v. 15, n. 4, pp. 293-301.

Emilia, S., Monica, Z., Christophe, P., and Jean, K., 1999, “Actigraphy and Leg Movements During Sleep: A Validation Study”, Journal of Clinical Neurophysiology., v. 16, n. 2, pp.154-160.

Ancoli-Israel, S., Cole, R., Alessi, C., Chambers, M., Moorcroft, W. and Pollak, C.P., 2003, ”The role of actigraphy in the study of sleep and circadian rhythms”, Sleep, v. 26, n. 3, pp. 342-92.

Ajilore, O., Stickgold, R., Rittenhouse, C.D. and Hobson, J.A., 1995, “Nightcap: laboratory and home-based evaluation of a portable sleep monitor”, Psychophysiology, v. 32, n. 1, pp.92-98.

Sadeh, A., Hauri, P.J., Kripke, D.F. and Lavie, P., 1995, “The role of actigraphy in the evaluation of sleep disorders”, Sleep, v. 18, pp. 288-302.

Choi, B H., Seo, J. W., Choi, J. M., Shin, H. B., Lee, J. Y., Jeong, D. U. and Park, K. S., 2007, “Non-constraining sleep/wake monitoring system using bed actigraphy”, Medical and biological engineering and computing, 2007, v. 45, n. 1, pp. 107-14.

Kristo, D. A., Eliasson, A. H., Poropatich, R. K., Netzer, C. M., Bradley, J. P., Loube, D. I. and Netzer, N. C., 2001, “Telemedicine in the Sleep Laboratory: Feasibility and Economic Advantages of Polysomnograms Transferred Online”, Telemedicine Journal and e-Health, v. 7, n. 3, pp. 219-224.

Seo, J., Choi, J., Choi, B., Jeong, D.U. and Park, K., 2005, “The development of a nonintrusive home-based physiologic signal measurement system”, Telemedicine Journal and e-Health, v. 11, n. 4, pp.487-95.

Choi, J.M., Choi, B.H., Seo, J.W., Sohn, R.H., Ryu, M.S., Yi, W., and Park, K.S., 2004, “A system for ubiquitous health monitoring in the bedroom via a bluetooth network and wireless LAN”, Proceedings of the IEEE 26th Annual International Conference on Engineering in Medicine and Biology Society, v. 2, pp. 3362-3365.

Hsu, Y. L., Yang, C. C., Tsai, T. C., Cheng, C. M. and Wu, C. H., 2007, “Development of a Decentralized Home Telehealth Monitoring System”, Telemedicine and e-Health, v. 13, n.1, pp. 69-78.

Chokroverty, S., Hening, W.A., and Walters, A.S., 2003, Sleep and Movement Disorders, Philadelphia, Elsevier Science.

Jean-Louis, G., von Gizycki, H., Zizi, F., Fookson, J., Spielman, A., and Nunes, J., 1996, “Determination of Sleep and Wakefulness With the Actigraph Data Analysis Software(ADAS)”, Sleep, v. 19, n. 9, pp. 739-743.

Cole, R.J., Kripke, D.F., Gruen, W., Mullaney, D.J. and Gillin, J.C., 1992, “Automatic sleep/wake identification from wrist activity”, Sleep, v. 15, n. 5, pp. 461-469.

Sadeh, A., Hauri, P. J., Kripke, D. F., and Lavie, P., 1995, “The role of actigraphy in the evaluation of sleep disorders”, Sleep, v. 18, pp. 288-302.

Rachwalski, T., Irvine, S., Steeper, J.W.T., Inkster, D.R., and Wells, C., 2005, “Poor quality of Sseep is a precursor of mobility-related adverse events”, Proceedings of Evidence-based strategies for Patient Falls and Wandering 2005.