//Logo Image
「世大智科/天才家居」-我們創業囉
PDF Version

作者:劉育瑋 (2014-07-17);推薦:徐業良(2015-12-11)

附註:本文為103學年度元智大學機械工程研究所劉育瑋博士論文「結合軟質活動感知床墊建立以床為核心之遠距居家照護系統」第四章。

Chapter 4. The bed-centered telehealth system for nursing applications

This chapter describes the Bed-Centered Telehealth System (BCTS) information structure, technical details and applications in nursing homes. By integrating the motion sensing mattress with a Care Management System (CMS), the BCTS demonstrates practicality in nursing homes and care homes.

4.1    Development of the care management system based on the motion sensing mattress

For enhancing the care quality of nursing homes with the existing manpower and care system, technology intervention for a more efficient management of care services has become an important issue in healthcare.

Generally, the bed plays a kernel role for the older adults living in the nursing homes or care homes. In order to ensure safety and health of the nursing home residents, the nursing staff has to visit them bed by bed to provide necessary aids and primary nursing care on schedule. Therefore, the bed is often used as a basic unit for care service management, as well as residents’ life and health management in nursing homes.

In order to meet the healthcare demands in nursing homes, a thorough investigation and discussion with the nursing and managing staff were conducted. From the practitioners’ needs and expectation, the user scenario of the CMS in the nursing home is defined as follows:

Ÿ   The nursing staff can monitor the bed-status of all residents in real time. While the resident leaves the bed, the bed-status changes from in-bed to off-bed. Then this information is transmitted immediately to the remote server. The information board of the CMS prompts the nursing and managing staff by means of visual or auditory messages, and the nursing staff is able to provide active service in time.

Ÿ   As the resident lies on the bed, the CMS starts to measure the elapsed time. If there is no body movements detected on the bed for more than two hours, a reminder is transmitted immediately to the information board of the CMS. This is important for disabled residents who cannot leave beds. With visual and auditory prompting messages, the nursing staff is then notified to visit the resident, help to relieve the pressure and prevent against complications such as bedsores by turning the resident’s body over and patting his/her back.

Ÿ   The long-term record of on/off bed and in-bed activity data collected by the CMS can be used as an important reference of the life patterns and sleep conditions of the residents.

Following these needs, the CMS based on the motion sensing mattress (WhizPAD) was developed for the applications in nursing homes. The infrastructure of the care management system is shown in Figure 4-1. The WhizPAD helps to collect the bed-activity signals and transmit them to the CMS. The system enables bed-status alert, reminder of resident visit, real-time and long-term record of in-bed activities monitoring. Following the messages displayed on the information board of the CMS, the nursing staff can keep aware of the residents’ needs for emergency alert and urgent care, as well as providing life and health management for them.

Figure 4-1. The infrastructure of the care management system

Figure 4-2 shows the information structure of the BCTS in the nursing home setting. The bedside processor (BDP) that accompanies the WhizPAD serves as an end device of a Zigbee wireless sensor network established in the nursing home. The monitoring data for each resident is transmitted directly to the remote server by a coordinator of the Zigbee wireless sensor network for the data management and service administration. Intermediate Zigbee routers can be deployed if the distances between end devices and the coordinator are too far. Figure 4-3 shows the data received by the remote server, including bed status, sleep posture, movement in bed, and elapsed time (minute) without movements.

Integrated with the CMS, the data received from the BDP are displayed on the information board at the local nursing station to facilitate real-time monitoring and alerts, service reminders, and browsing the historical data record. Figure 4-4 shows the main interface of the information board. The nursing staff can keep aware of whether a resident is on the bed, as well as when to visit disabled residents who cannot leave beds, to help turn the resident’s body over or pat his/her back. The nursing staff can also query the data for a particular resident from their mobile devices. Physical activities in bed and classified events are stored in the historical database and can be used not only in the management of the particular resident but for administrative purposes such as ensuring that adequate staff is on duty. Figure 4-5 shows the screen of daily record of monitoring data. The nursing staff can browse the on/off bed status, sleep posture and number of body movement in bed of a resident on a certain day.

System structure of WhizPAD in nursing home

Figure 4-2. The Bed-Centered Telehealth System in a nursing home scenario

Figure 4-3. The interface of data receiving on remote server

圖片13r23r23r3

Figure 4-4. The main interface of the information board at the nursing station

Figure 4-5. The individual historical data screen

4.2    System implementation

The CMS based on WhizPAD has been implemented in Yi-De Nursing Home in Taoyuan, Taiwan. Figure 4-6 shows the map of the CMS installation in Yi-De Nursing Home. There are 30 WhizPADs putting atop residents’ beds in five rooms (R2210, R2220, R2310, R2320 and R2330).

Figure 4-7 shows the setup of WhizPAD in residents’ room. WhizPAD is covered with a blue water-poof cover for preventing bed-wetting events. The BDP stuck on the wall is powered by a +5 volts adapter, then the WhizPAD can work normally. The coordinator of Zigbee wireless sensor network in nursing station receives the monitoring data from WhizPADs. By connecting to the internet access point, the coordinator also transmit the data to the remote server through internet for data storage and management. Seven Zigbee routers are used to help the monitoring data transmission from residents’ room to coordinator of the Zigbee wireless sensor network, as shown in Figure 4-8. The user interface of the CMS is installed in six AIO computers connecting to the remote server, to display the monitoring data and abnormal event alerts. As shown in Figure 4-9, the AIO computers are put in the nursing station and each care room for immediate care service. These computers are also easily operated by touch-screens.

Figure 4-6. The map of the CMS installation in Yi-De nursing home

Figure 4-7. The setup of WhizPAD in residents’ room

Figure 4-8. The setup of Zigbee coordinator and routers

Figure 4-9. AIO computers in nursing station and each care room for immediate care service

Figure 4-10 and 4-11 are the sample data of residents collected by the BCTS in the nursing home. These figures show bed related activities of four residents with different conditions in a typical day, including the on/off bed status and the number of movements in bed per minute. Two bed related indices, the average number of body movements in bed per minute and the percentage of in-bed time in a day are also displayed. Figure 4-10 (a) shows the data of a healthy resident in a typical day, and Figure 4-10 (b) shows the data of a resident with frequent body shaking and twitching. Both residents have very similar life pattern but the average number of body movements in bed per minute for the resident in Figure 4-10 (b) is much higher. Figure 4-11 (a) shows the data of a disabled resident who cannot leave the bed. There are intense movements in bed in regular periods (about every two hours), which are actually the care services of body turning over, to relieve the pressure and prevent complications such as bedsores. Figure 4-11 (b) shows a completely different data pattern obtained from a dementia resident.

Figure 4-10. Data collected by the BCTS: (a) a healthy resident, and (b) a resident with frequent body shaking and twitching

Figure 4-11. Data collected by the BCTS: (a) a disabled resident, and (b) a dementia resident

4.3    Interpretation of physical activities in bed of residents

Based on the monitoring of long-term physical activities in bed, the system can help the nursing staff to realize the life patterns and sleep conditions of residents in the nursing homes. For interpreting the long-term physical activities in bed of residents in the nursing homes, two groups of residents, five residents with motor capability and five residents without motor capability were recruited. Continuous data of one week of physical activities in bed was collected with the consent of the ten participants (three males, seven females, aged 65-87 years old, weighing 38-73 kg). In this investigation, three bed-related indices were used to interpret the long term physical activities in bed of residents, including the average number of body movements in bed per minute (/minute), percentage of in-bed time in a day (%) and off-bed times in a day (/day).

4.3.1 Data interpretation of residents with motor capability

The one week data collcted from the five residents with motor capability was analyzed to interprete their in-bed pattern in the nursing home. Table 4-1 shows the analysis result of their bed-related indices. The time in bed of the five residents are about 50% or more; the average number of movements per minute in bed ranges from 0.31 to 1.08; the average number of getting off-bed ranges from 1.86 to 8.14. Figure 4-12 shows the long-term data of a normal resident with a regular in-bed pattern (Case 3). The in-bed time in a day is regular, only the body movement in bed is different. Figure 4-13 shows the long-term data of a normal resident (Case 5) with frequently getting off-bed events. This resident has the highest average off-bed time in a day (8.14) among the 5 residents.

Table 4-1. Three bed-related indices for the five residents with motor capability

Case

Bed-related indices

D1

D2

D3

D4

D5

D6

D7

Average/SD

1

Movements in bed

0.22

0.19

0.18

0.22

0.52

0.54

0.31

0.31/0.15

Time in bed (%)

55.34

55.58

59.59

56.57

59.32

54.80

56.87

56.87/1.91

Off-bed per day

2

3

1

2

3

2

3

2.29/0.76

2

Movements in bed

0.28

0.24

0.20

0.39

0.36

0.61

0.39

0.35/0.14

Time in bed (%)

60.79

62.93

64.33

50.19

59.66

45.42

53.59

56.70/7.09

Off-bed per day

2

3

3

1

2

3

1

2.14/0.90

3

Movements in bed

1.02

0.11

0.40

0.48

0.12

0.23

0.23

0.37/0.31

Time in bed (%)

42.60

47.32

47.94

46.87

48.45

47.50

55.21

48.04/3.71

Off-bed per day

1

1

1

1

1

1

1

1.00/0.00

4

Movements in bed

1.50

1.44

0.28

0.52

0.29

0.18

1.22

0.78/0.59

Time in bed (%)

47.41

47.66

47.83

58.75

48.74

66.53

61.29

54.10/8.07

Off-bed per day

1

1

1

2

1

3

4

1.86/1.21

5

Movements in bed

1.19

0.57

1.18

0.79

1.50

1.52

0.84

1.08/0.36

Time in bed (%)

57.08

64.94

57.83

60.14

60.67

57.00

62.46

60.01/2.97

Off-bed per day

9

9

7

8

9

8

7

8.14/0.9

Figure 4-12. The long-term data of a normal resident with a regular life pattern (Case 3)

Figure 4-13. The long-term data of a normal resident with frequently getting off-bed events (Case 5)

4.3.2 The data interpretation of residents without motor capability

The one-week long term data collcted from the five residents without motor capability was analyzed for exploring their in-bed patterns in the nursing home. Table 4-2 shows the analysis result. These residents spent more than 90% time in a day lying on the bed. The average number of movements in bed of disabled residents ranges in 0.54 to 1.13. These movements in bed causes by the care service of turning the body over and patting the back provided by nursing staff. The nursing staff also helped these residents to leave the bed for care services such as taking a shower. The average number of getting off-bed events ranges from 0.43 to 1.14, which is lower than the residents with motor capability. Figure 4-14 shows the long-term data of a disabled resident with the care services of body turning over, to relieve the pressure and prevent the bedsores event.

Table 4-2. Three bed-related indices for the five residents without motor capability

Case

Bed-related indices

D1

D2

D3

D4

D5

D6

D7

Average/SD

6

Movements in bed

1.30

0.53

1.12

0.62

1.00

0.94

0.99

0.93/0.27

Time in bed (%)

100

99.73

99.06

100

98.50

100

100

99.61/0.59

Off-bed per day

0

1

1

0

2

0

0

0.57/0.79

7

Movements in bed

0.39

0.58

0.67

0.79

0.74

1.03

0.34

0.65/0.24

Time in bed (%)

99.18

100

100

98.71

100

98.39

100

99.47/0.70

Off-bed per day

1

0

0

1

0

1

0

0.43/0.53

8

Movements in bed

0.68

0.75

0.51

0.51

0.52

0.50

0.65

0.59/0.1

Time in bed (%)

99.48

100

99.04

100

98.27

100

99.43

99.46/0.64

Off-bed per day

1

0

1

0

1

0

1

0.57/0.53

9

Movements in bed

0.63

1.16

1.56

1.35

1.02

1.33

0.85

1.13/0.32

Time in bed (%)

100

75.16

96.10

69.00

78.20

94.79

93.75

86.71/12.26

Off-bed per day

0

2

0

2

1

1

2

1.14/0.90

10

Movements in bed

0.93

0.76

0.34

0.88

0.71

1.51

1.69

0.97/0.47

Time in bed (%)

100

89.96

100

100.0

100.0

77.54

83.68

93.01/9.42

Off-bed per day

0

2

0

0

0

2

2

0.86/1.07

 

Figure 4-14. The long-term data of a disabled resident with the care services of body turning over, to relieve the pressure and prevent the bedsores event (Case 9)

4.4    Long-term in-bed patterns of residents

Applying the average profile of physical activities in bed decribed in Chapter 3, the in-bed pattern of residents in the nursing home can be found. By calculating the correlation between the average profile and the monitoring data, the correlation coefficient can provide the reference for the nursing staff to realize whether the residents in nursing home have good in-bed pattern.

Figure 4-15 shows examples of correlation coefficients between the one-day monitoring data and average profiles of Case 3, 5 and 9. The red line represents the seven-days average profile of in bed status, and the monitoring data of given day (the 7th Day) is the blue line. The correlation coefficients of in-bed status are 0.89, 0.80, and 0.92, respectively. Based on the analysis of long-term in-bed patterns, the disabled resident (Case 9) who lie on the bed almost all day long has the higher correlation coefficient (0.92); and normal resident (Case 5) with frequently off-bed events has the lower correlation coefficient (0.80).

Figure 4-15. Examples of correlation coefficient between the one-day monitoring data and average profiles (a). Case3 (b). Case5 (c). Case 9