Proposal.doc

Proposal

SMART MATTRESS SYSTEM FOR PATIENT IDENTIFICATION

AND BEDSORE PREVENTION

ECE4007 Senior Design Project

Section L03, Smart Mattress Team

Bryan KuoPriyen PatelDev ShahXitij Shah

Tim Stamm

SubmittedSeptember 15, 2008

Smart Mattress (ECE4007L03)

TABLE OF CONTENTS

Executive Summary.......................................................................................................... iii

1. Introduction....................................................................................................................11.1 Objective ...........................................................................................................11.2 Motivation .........................................................................................................11.3 Background .......................................................................................................2

2. Project Description and Goals .....................................................................................2

3. Technical Specification..................................................................................................3

4. Design Approach and Details4.1 Design Approach ...............................................................................................44.2 Codes and Standards........................................................................................114.3 Constraints, Alternatives, and Tradeoffs .........................................................11

5. Schedule, Tasks, and Milestones.................................................................................12

6. Project Demonstration..................................................................................................13

7. Marketing and Cost Analysis7.1 Marketing Analysis...........................................................................................137.2 Cost Analysis ...................................................................................................14

8. Summary........................................................................................................................16

9. References......................................................................................................................17

Appendix A........................................................................................................................19

ii


EXECUTIVE SUMMARY

Infections acquired in hospitals cost the healthcare industry billions of dollars each year

and result in numerous preventable deaths. These infections can result from patients developing

bedsores from lying stationary for an extended period or wetting the bed. In other instances, the

patient in the bed is mistaken for the wrong patient and given incorrect medication or treatment.

To prevent these overlooked problems, the Smart Mattress intends to verify the correct patient is

occupying the bed, detect patient movement, and sense the presence of moisture.

Proper patient identification will be assured by displaying the patient’s name and

medication ID bar code on a PC monitor using RFID to read data from the patient’s ankle tag.

An array of pressure sensors will be used to monitor patient movement and warn staff when a

patient needs to be moved. To detect moisture, a disposable mattress cover will be created using

a conductive pattern between the top paper layer and the bottom biodegradable plastic layer.

Since polymer reinforced paper loses its strength when wet, air-laid paper widely used in

hospital applications can be used to provide a safe, flexible, and low cost alternative.

The Smart Mattress will be demonstrated in a classroom of the Van Leer building located

on the Georgia Institute of Technology main campus. The functionality of the mattress will then

be tested by simulating several possible scenarios that could occur in a typical patient room.

Specifically, a test “patient” will go through scenarios that test the functionality of the wetness

detection system, the movement detection system, and the patient identification system.

When the system is mass-produced, the cost of each unit will be $1,305. With the system

being so cheap, a profit margin of 15 percent is expected with a final selling price of $1,505.

iii


1. INTRODUCTION

Nosocomial infections, which are infections acquired in hospitals are prevalent at a rate

of about 9.8 percent per every 1000 days that a patient spends in the hospital. These infections

cost the healthcare industry $4.5 billion and caused 88,000 deaths in 1995 [1]. These infections

can result from patients developing bedsores from lying stationary for an extended period or

wetting the bed. In other instances, the patient in the bed is mistaken for the wrong patient and

given incorrect medication or treatment. To prevent these overlooked problems, the Smart

Mattress intends to verify the correct patient is occupying the bed, detect patient movement, and

sense the presence of moisture.

1.1 Objective

The purpose of the project is to design an inexpensive mattress system that prevents bed-

related nosocomial infection and identifies occupants. The mattress will be able to detect

occupant movement. The mattress cover will be able to detect the presence of moisture.

Pressure and moisture sensors will be connected to a system that will notify hospital staff in the

event of a problem. The mattress will also identify the patient and display the patient’s name and

barcode on a monitor for easy medicine distribution. This product will be marketed to hospitals,

clinics, and nursing homes.

1.2 Motivation

According to the CDC, 2 million people become infected with nosocomial infections

every year. Increasingly, patients are contracting MRSA infections, which are harder to treat

with antibiotics, as they are more resistant than normal bacterial infections [2]. Infection rates

vary between hospitals and nursing homes, but on average, 10 percent of hospital patients

contract an infection, while the average rate at nursing homes is 25 percent [3].

1


Though other smart beds are available to hospitals, these beds have many extra features

not required by hospitals that already have equipment to perform these functions. The Smart

Mattress will operate using stand-alone devices to be cost-effective.

1.3 Background

Several different designs for smart beds are used to reduce the pressure on the patient [4].

The problem with the smart beds on the market is they are expensive and have only one use,

which is to prevent bedsores; they overlook bed-wetting, which is also a major source of

infection.

In recent years, expansive research in the commercial uses of RFID has led to major

advances in the healthcare industry. The market for RFID tags and systems in healthcare is

expected to grow to $2.1 billion by 2016 [5]. In the hospital environment, it is important for the

doctor to be able to access a patient’s information as fast as possible in case of an emergency, or

to make sure that the correct patient is being treated.

2. PROJECT DESCRIPTION AND GOALS

The Smart Mattress will seamlessly integrate patient identification and bed sore prevention

into a typical hospital bed. The product will do the following:

Identify patients using RFID

Display patient information and barcode on a PC monitor

Detect moisture and pressure that could create bed sores

Alert staff if patient is in danger of developing bed sores

The Smart Mattress will be equipped with a RFID receiver to communicate with ankle tags worn

by the patient. Once the patient is identified, an external PC monitor will display the name of the

patient and a barcode that when scanned will display to hospital staff the correct medications to

2


administer. Pressure and wetness sensors will also be located in the Smart Mattress to detect the

presence of fluid and to detect pressure. If the patient is in danger of developing bedsores, the

hospital staff will be alerted. The expected unit price of the Smart Mattress is $1,305, which is

cost-effective in the target market of hospitals and nursing homes.

3. TECHNICAL SPECIFICATIONS

The following table shows the proposed technical specifications for the technology used

to design the Smart Mattress. A maximum RFID antenna detection range of twelve inches is

proposed for the tags. This range is sufficient for the purposes of detecting the patient lying on

the bed.

Table 1. Technical Specifications

The antennas for the RFID detection system will be placed between two-inch thick foam

mattresses. The placement of RFID antennas in between the two foam mattresses and near the

bottom of the bed will provide the best orientation for detecting a tag worn by the patient around

the ankle. The 1W transmitting power secures that the twelve-inch range is detectable. The

frequency of 13.56 MHz was chosen due to the commonality of equipment at this specification.

3

Technical Specifications ProposedRFID Reader RFID Operating Frequency 13.56 MHzRFID Transmitting Power 1W ± 2 dBMax RFID Reading Distance 1 ft (12 in.)RFID Current Draw Max. 0.5 AMultiplexer Microprocessor PIC18 LF2321Power Supply 3.3 VCurrent Consumption < 5mASwitching Loop-time 16 msTrossen Robotics 24” FSR Force Sensitivity Range <100 g to > 10 kgPressure Sensor Range < 1.5 psi to > 150 psi


This high of a frequency is required as not to interfere with or be interfered by other hospital

equipment.

The movement monitoring system is implemented using four Force Sensing Resistor

(FSR) strips to measure the pressure applied by the patient to the mattress. The pressure strips

will be placed between the two foam mattresses. The sensitivity range of the pressure strips can

be modified by varying the feedback resistor. Modifying the pressure range allows the pressure

strips to measure up to 1,000 lbs.

4. DESIGN APPROACH AND DETAILS

4.1 Design Approach

The Smart Mattress project is comprised of patient identification and bedsore prevention

features. Patient identification will be accomplished using an RFID system, and bedsores will be

prevented by coupling movement and wetness monitoring systems. Each system will be

implemented separately using previous groups’ design solutions, and then modified in order to

combine them into a full product package.

Patient Identification

Proper patient identification will be assured by displaying the patient’s name and

medication ID bar code on a PC monitor using RFID. The RFID system will have four

components: active RFID tags, antenna receiver, a 13.56 MHz transceiver, and a patient ID

database.

Each tag has a unique ID corresponding to the tag’s owner. The RFID transceiver signals

the tag through an antenna. Upon receiving the transmitted signal from the antenna, the inner

circuitry in the tag returns the unique signal representing the binary ID assigned to each patient.

This response signal is then detected by the antenna, decoded in the transceiver, and then

processed on a PC.

4


Once a tag is detected by an antenna, the serial number of the tag is compared to a patient

ID database on the PC. In this database, information such as the patient’s name will be stored.

A bar code will be generated based off the patient’s unique ID. The patient’s name and unique

bar code will then be displayed on a PC monitor to help secure proper medication delivery.

Bedsore Prevention

An array of pressure sensors will be used to monitor patient movement and warn hospital

staff when a patient needs to be moved. Movements will be monitored using the algorithm

shown in Figure 1. The movement monitoring system will serve three main purposes. The

system will detect the patient’s weight to check if there is a patient in the bed. Preventing

bedsores will be achieved by detecting significant movements by the patient; a significant

movement will be defined as 20 percent change from the pressure sensors’ last reading. Once a

significant movement is detected, the thirty-minute timer will be reset. If the patient does not

move within thirty minutes, the LED alarm will be set off.

5


Figure 1. Movement detection algorithm.

The pressure sensing will be carried out by several force sensing resistors (FSR) from

Trossen Robotics. The FSR functions by decreasing its resistance with an increase in force. The

sensitivity of the FSR can be adjusted using a voltage divider, which is then interfaced with a PC

using the Trossen Robotics Interface Kit; all three of these components are shown in Figure 2.

6


Figure 2. (a) TR Voltage Divider, (b) TR Phidget Interface Kit, (c) Force Sensing Resistor.

Figure 3 shows how the FSRs will be laid out in relation to the patient and the mattress.

Figure 3. Pressure sensors on Smart Mattress.

The middle area of the bed is the best spot to put the pressure sensors because this is the area of

the body where bedsores are most likely to develop [6]. The FSRs will be placed in the bed so as

not to allow the sensors to bend.

7


Wetness Detection System

A disposable mattress cover will be created using a conductive pattern between the top

paper layer and the bottom biodegradable plastic layer as designed by a previous group. Since

polymer reinforced paper loses its strength when wet, air-laid paper widely used in hospital

applications can be used to provide a safe, flexible, and low cost alternative. A conductive

pattern will be interfaced with a wetness detection circuit using comparators to prevent the

patient from remaining in a wet bed. Combined with the movement monitoring system, this will

greatly reduce the chances of the patient developing a bedsore.

Conductive Pattern

A conductive pattern similar to Figure 4 will be constructed for use inside the mattress

cover. Two circuit loops separated with a 6” gap are used for signals of different magnitude.

The detection circuit interfaces with the circuit loops via the four leads at the edge of the

mattress cover. Wetness can be determined by

measuring the resistance across the two loops. The

resistance of wet paper is significantly lower than

dry paper since urine and blood have a high

concentration of electrolytes. Broken loops are

also detected by measuring the resistance of the

loop. The state of the mattress cover is displayed

using two LEDs. One LED will light to indicate the

mattress cover is wet and the other to indicate a

broken loop.

Figure 4. Wire layout on mattress sheet.

8


Wetness and Continuity Detection

The detection circuit is shown in Figure 5. Resistors R2 and R4 model the conducting

loops found on the mattress cover. AC current is divided into two signals for comparison, one of

which is halved through voltage division by R8 and R9. Three LM311 comparators detect

wetness and breakage of loops, signaling breakage by turning off LEDs X1 and X2 and signaling

wetness by turning on X3. Mattress cover designs with different resistances can be accounted

for by tweaking resistor values.

PIC Microcontroller

A microcontroller-based design will reduce costs, lower current consumption, increase patient

safety, and provide a much more compact solution than a PC. Integrating patient identification

and bedsore prevention features into one compact product will be accomplished using a

PIC18LF2321 microcontroller as seen in Figure 6.

Figure 5. Wetness and broken lead detection circuit.

9


Figure 6. Pic18LF2321 Microcontroller and Pin Layout.

The microcontroller will be used in place of a PC to control and process data from the

RFID transceiver, FSRs, and wetness detection circuit. Using the onboard UART (universal

asynchronous receiver transmitter) communication between the PIC and RFID transceiver will

be accomplished via the RS232 standard. The FSRs will be monitored using the onboard ADC

and four of the PIC’s analog inputs. Wetness detection circuit outputs will be tied to digital input

pins on the PIC instead of LEDs.

Power saving requires the partitioning of the design based on power consumption during

its operation by determining the required operation states and shutting down unwanted circuitry.

Idle modes power down the CPU while allowing peripherals such as an ADC to continue to

operate. Most power conserving applications need the system controller to remain in a low

power state most of the time, waking up periodically under a timer’s interrupt to run program

code. By invoking the sleep state, the power consumption can be reduced by as much as 96

percent [7].

10


4.2 Codes and Standards

The RFID technology implemented in the Smart Mattress system must meet the stringent

codes and standards of typical hospital equipment. RFID antennas pose a threat to patients if they

transmit a high-powered signal near patients. To minimize the power radiated by the RFID

antenna, the distance between the antenna and its receiver must also be minimized. Several tests

will be performed to determine the appropriate proximity of the antenna with respect to the

receiver in order to avoid harm to medical patients occupying the Smart Mattress.

Furthermore, the RFID system must meet electromagnetic compatibility (EMC)

standards to prevent the interference with other significant medical equipment [8]. Specifically,

the RFID system must meet the standards specified by SC 31, which state that “device

manufacturers claiming conformance to this standard shall self-certify that RF emissions and

susceptibility comply with IEC 60601-1-2” [9]. The operating frequency of the RFID system

integrated in the Smart Mattress is 13.65 MHz; therefore, it meets the EMC requirements for

medical devices.

4.3 Constraints, Alternatives, and Tradeoffs

Alternatives to moisture detection systems exist, while complete wetness detection

systems requiring no assembly can be purchased. These systems often come with detailed

instructions and software to interface with a PC. However, a moisture detection system built by

the team will cost less because it can be created using cheap components that are easily

obtainable. This system can also be modified to interface with different hardware components.

Because cost and flexibility are important considerations in this project, the team-built moisture

detection system offers more advantages than a retail system.

11


Retail RFID antennas present a similar situation. Buying an antenna decreases the

amount of time needed to design and build the patient identification system. However, antennas

available for purchase cost more than the allotted budget. Although building an antenna is a

lengthy process, the time needed to do so would not cause missed deadlines. Keeping the project

within cost constraints is more significant than reducing design and manufacturing time.

5. SCHEDULE, TASKS, AND MILESTONES

The prototype of the Smart Mattress will be built according to the timeline given below.

The team is divided into two subgroups that work concurrently. Bryan, Priyen, and Xitij will

focus on the bedsore prevention system. Dev and Tim will concentrate on the patient

identification system. The subgroups meet several times a week to work on tasks, and the

complete team meets at least twice a week with the project adviser to discuss progress, problems,

and solutions. Both subgroups separated the tasks into subtasks to define responsibilities. These

are shown in the Gantt chart in Appendix A.

Construction of the Smart Mattress will begin on September 22. The hardware will be

installed first. While the PIC is being programmed, the detection and identification systems will

be set up and implemented. By October 13, both systems will be ready to be tested and debugged

Table 2. Project Timeline

WBS

Task Name Duration

Start Finish Difficulty

Responsible Person

1 Define Project 21 days 8/18/2008

9/15/2008 Easy All Members

2 Acquire Parts 5 days 9/15/2008

9/19/2008 Easy All Members

3 Implement Microcontroller 11 days 9/22/2008

10/6/2008 Medium Tim

4 Implement Bed Sore Detection 29 days 9/22/2008

10/30/2008

High Bryan, Priyen, Xitij

12


independently. The full system will be assembled and will be ready for demonstration by

November 27.

6. PROJECT DEMONSTRATION

The Smart Mattress will be demonstrated in a classroom of the Van Leer building located

on the Georgia Institute of Technology campus. The team will initially give a power point

presentation explaining the background information, general functions and underlying

technology of the Smart Mattress. The functionality of the mattress will then be tested by

simulating several possible scenarios that could occur in a typical patient room. Specifically, a

test “patient” will go through scenarios that test the functionality of the wetness detection

system, the movement detection system, and the patient identification system. For example, to

test the wetness detection system, a team member will pour salt water on the mattress. If the

system detects the wetness, an LED will turn on to notify a nurse that the mattress is wet. The

“patient” will also simulate cases that will test how these different systems (wetness detection,

pressure detection, RFID, and PC monitor) interact with each other.

7. MARKETING AND COST ANALYSIS

7.1 Marketing Analysis

Many healthcare companies have developed “intelligent” patient beds in response to the

increasing occurrence of nosocomial. For example, TRADEWIN manufactures an Alternating

Pressure Mattress System that monitors pressure distribution and modifies airflow to help

prevent bedsores. The TRADEWIN 3000 8” Alternating Pressure Mattress system is sold for

$1,350 per unit [10]. MED-AIRE also manufactures a similar alternating pressure mattress for

$1,099 per unit [11]. Although the Smart Mattress system is more expensive than the previously

mentioned competitors, approximately $1,505 per unit, it possesses a unique set of features that

13


distinguish it from any intelligent hospital bed in the market.

The Smart Mattress system differs from many intelligent patient beds in the health care

industry because of its versatility. The proposed mattress system not only measure pressure, but

also measure wetness, which is a major contributing factor in the development of bed sores.

Along with bed sore prevention technology, the Smart Mattress also utilizes an RFID system to

display the patient’s name on a PC monitor located near the hospital bed. The monitor will also

display important medication information to prevent any potential medicine distribution errors.

7.2 Cost Analysis

Table 3 shows the cost of the all the parts that will be used to make the Smart Mattress

system. The grand total of the parts to make one Smart Mattress system is $770. The most

expensive component is the RFID system, which includes the reader and the transmitter.

Table 3. Part Costs

Part Quantity Unit Cost Total Cost

FSR Robotics Sensor Kit 2 $27 $54

PIC18LF4321 1 $6 $6

Wires/Cables 1 $15 $15

Button Cell Battery 1 $15 $15

RFID Tag 1 $7 $7

RFID Reader 1 $300 $300

Conductive Fabric Tape 100ft $80 $80

Mattress 1 $150 $150

PIC Programmer Kit 1 $90 $90

Bed Sheet 1 $3 $3

PC Monitor 1 $50 $50

Total Equipment Cost $770

14


Table 4 shows a list of the costs of developing the Smart Mattress system. Including the

labor to develop the system, the total cost of the initial Smart Mattress system is $34,000. The

labor cost was estimated based on the starting salary of $52,200 for someone who graduated

from the Georgia Institute of Technology with an electrical engineering degree [12]; this equals

$26.10/hour based on the 40-hour workweek. Fringe benefits and overhead were calculated at 25

percent each.

Table 4. Development Costs

Component Labor

Hours

Labor

Cost

Total

Component

Cost

Pressure Sensor Testing 150 $3,915$9,135

Wetness Detection Testing 200 $5,220

RFID Sensing 100 $2,610$5,740

RFID PC Monitor 120 $3,130

Design Meetings 240 $6,265 $6,265

Total Labor 810 $21,140

Total Equipment Cost $770

Fringe Benefits, 25% Of Labor $5,285

Overhead, 25% Of Equipment,

Labor & Fringe

$6,800

Total Overhead $12,080

Total Project Cost $34,000

Table 5 shows the projected costs and revenue of when the Smart Mattress system is put on the

market. When the system is mass-produced, the cost of each unit will be $1305. With the system

being so cheap, a profit margin of 15 percent was used for a final selling price of $1,505. The

15


projected rate of sales over five years is 15,000 units, for total revenue of $22,575,000 and a total

profit of $3,386,250.

Table 5. Profit Projection

Equipment Cost $770

Assembly Labor $12

Testing Labor $8

Subtotal, Labor $20

Fringe Benefits, 25% of Labor $5

Subtotal, Labor & Fringe $795

Overhead, 25% of Material, Labor & Fringe $200

Subtotal, Input Costs $995

Sales & Marketing Expense, 20% of Production Price $200

Support & Warranty Expense, 10% of Production Price $100

Amortized Development Costs $10

Subtotal, All Costs $1,305

Profit, 15% $200

Selling Price $1,505

Total Revenue, Based on 15,000 Units over 5 years $22,575,000

Total Profit $3,386,250

8. SUMMARY

The “Smart Mattress” design group is currently finished with determining which features

to add to the patient bed. These features include identifying patients using RFID, displaying

patient information and barcode on a PC monitor, detecting moisture and pressure that could

create bedsores, and alerting a staff member if a patient is in danger of developing bedsores.

Once the most cost effective design is formulated, the group will then conduct research and

testing on the RFID and bed sore detection systems.

16


9. REFERENCES

[1] R.A. Weinstein. (July, 1998). Nosocomial Infection Update. Emerging Infection

Diseases [Online]. 4(3). [cited 2008 Sep 11], Available:

http://www.cdc.gov/ncidod/eid/vol4no3/weinstein.htm

[2] R. Moser, “Dirty Places, Part 12: Hospitals/Nursing Homes,” [Online Document],

[cited 2008 Sep 11], Available:

http://blogs.webmd.com/all-ears/2006/08/dirty-places-part-12-hospitalsnursing.html

[3] M. Haggerty, Gale Encyclopedia of Medicine, “Bed Sores” [Online], [cited 2008 Sep 11], Available:

http://www.healthatoz.com/healthatoz/Atoz/common/standard/transform.jsp?requestURI

=/healthatoz/Atoz/ency/bedsores.jsp.

[4] R&D Products, LLC, “The Smart Bed,” [Company Website], [cited 2008 Sep 12],

Available: http://thesmartbed.com/products.htm

[5] A. Lewcock, “Healthcare RFID market forecast at $1.2B,” [Online Document], 2007 July

09, [cited 2008 Aug 30], Available:

http://www.healthcareitnews.com/story.cms?id=7436

[6] P. Chung, Y. Hur, M. Wozniak, D. Yoon, "Disposable Mattress Cover with Wet Sheet

Sensors," [Online Document], [cited 2008 Sep 11], Available:

http://www.ece.gatech.edu/academic/courses/ece4007/08spring/ece4007l05/ak15/

proposal.pdf

[7] J.B. Peatman, Coin-Cell-Powered Embedded Design, Atlanta, GA: Quick&Low Books,

2008, pp. 7-14

[8] W. Khawaja, M. Saleheen, S. Sanyal, B. Virk, and R. Eiswerth, “Intellibed Hospital Bed

Add-On Kit For Improved Patient Safety,” [Online Document], [cited 2008 Sep 11],

Available: http://www.ece.gatech.edu/academic/courses/ece4007/08spring/ece4007l05/

ak11/files/Proposal.pdf

17

http://www.ece.gatech.edu/academic/courses/ece4007/08spring/

http://www.healthcareitnews.com/story.cms?id=7436


[9] C. K. Harmon, “RFID: Update on Standards and Regulatory Initiatives,”

[Online Document], [cited 2008, Sep 12], Available:

http://www.aimglobal.org/members/news/templates/template.aspx?articleid=3302&zonei

d=45

[10] IDT Marketing, “Tradewind 3000 8” Alternating Pressure Mattress System,”

[Company Website], [cited 2008 Sep 12], Available:

http://www.alternatingpressuremattress.com/3000.html

[11] uCanHealth, “MED-AIRE 8” Alternating Pressure Mattress Overlay with Low Air Loss,”

[Company Website], [cited 2008 Sep 12], Available:

http://ucanhealth.com/goto.php?page=detail.php&graph1=14028&cat_page=alternating_

pressure_mattress

[12] Georgia Institute of Technology, “Bachelor’s Degree Candidates by Major,”

[Online Document], [cited 2008, Sep 12], Available:

http://career.gatech.edu/students/bachelor.pdf

18


APPENDIX A

GANTT CHART

19

20

Proposal.doc

Business

patient movement

test patient

patient identification

proposal smart mattress

smart mattress ece4007l0311

inexpensive mattress

correct patient isoccupying

proper patient identification