Towards Smart E-Health Sensor Networks: A Survey



Sensor Networks for E-Health: A Survey

Rabie A. Ramadan, Ahmed Y. Khedr, Alaa Hamouda, Ashraf Almrakby

Systems & Computer Department

Al-Azhar University in Cairo

Cairo, Egypt

Abstract

Health care – or sometimes referred to as E-health- has recently been a hot topic for researchers in computer science. The availability of new sensing technologies helps develop different types of health care applications and raised the concept of e-medicine as well. This paper surveys the state of the art of health care sensing technologies, devices, and applications. In addition, the use of expert systems in health care is presented. This includes diagnosis, suggestion for treatment, and prediction possible effects of different therapeutic interventions.  

 Keywords

Heath care, E-health, E-medicine, sensor network, health management, patient monitoring.

1. Introduction

Health care is an essential topic that includes many issues. Some of these issues are related to the patient health and his/her protection, others are related to the patient information, and some others are associated to the used technology. In fact, these issues cannot be totally separated. Patient health is the responsibility of the hospital staff that they might depend on smart software and hardware for diagnosis as well as monitoring the patient’s health; Artificial Intelligence (AI) and expert systems play major roles in such cases which is included as a core of many of the health care applications. At the same time, patient information is kept secure on a computer at the hospital. Staffs are the only persons that can access such information.

As can be seen from the previous scenario, there are many restrictions on such approach of handling patients’ information. For instance, if a patient in a critical health condition is admitted to another hospital that does not have his/her health record; such information might cost the patient his/her life. Nowadays, the advances of the sensing technology and networking lead to a revolution in the health care application. Electronic health (E-health) is named after the technology took place in the health applications. Patients are not required to stay at hospitals for continuous monitoring any more. They can stay at their comfort zones and their status is continuously reported to the hospitals where staff can see such information. Small but smart sensing devices such as pressure sensors play a critical rule in such case.

Throughout this paper, we give the reader the state of the art in the e-health technology and applications. Up to our knowledge, this is the first work that surveys the sensing technology along with e-health applications. We think that this survey will be beneficial to all of the researchers that are working on sensors, sensor networks, and e-health. It is the also, the first step towards deep investigation in e-health solutions.

The paper is organized as follows: Health Care Medical and Sensing Technology are investigated in the next section. Section 3 looks at some of the current sensors prototypes as well as commercial sensors that are used in the field of e-health. Section 4 presents the up-to-date e-health applications; some of these applications are very simple; others are very complicated in terms in sensing and networking. The current E-Health expert systems are categorized for the benefit of the reader in section 5. Section 6 gives many open problems in e-health. Finally, the paper concludes in section 7.

2. Health Care Medical and Sensing Technology

Sensing technology has come a long way. Many devices have been produced in the last few years due to the high demands on such technologies as well as their applications. For instance, there are many new sensors for location identification, e.g. GSM cell and Wi-Fi, proximity, e.g. RFID tags, movement, e.g. movement sensors, and health sensors, e.g. basic vital signs and glucose. A sensor is used to measure the physical environment such as temperature, pressure, glucose level, heart rate, etc. The signal measured by the sensor is transferred to the processing unit via I/O ports. After processing, and recording these signals, they can be sent to a remote server. Some devices can display this information to the patient via LCD displays.

In this section, we provide the reader with the state of the art sensing technology and sensors in the field of e-health including the commercial sensors. Such survey is intended to provide the functionalities of each sensing device as well as its applicability in fitting a real life sensor networks.

1. Sensors Architecture

Sensors are usually consists of four main units in addition to the two optional units which are mobility support and location finding units. The four main units are, as shown in Figure 1, sensing, processing, communication, and power units.

[pic]

Figure 1: Sensors Units

Sensing Unit

A sensor is used to convert a physical parameter into electrical signal. Most of these signals are analog and week signals. Therefore, the main components of the sensing unit are the sensors which differ from application to another. The second component is the analog to digital (ADC) converter. Depending on the application a variety of sensors are available. These sensors include but not limited to ‎[12]:

• an ECG (electrocardiogram) sensor for monitoring heart activity

• an EMG (electromyography) sensor for monitoring muscle activity

• an EEG (electroencephalography) sensor for monitoring brain electrical activity

• a blood pressure sensor

• a tilt sensor for monitoring trunk position

• a breathing sensor for monitoring respiration

• movement sensors used to estimate user's activity

• a "smart sock" sensor or a sensor equipped shoe insole used to delineate phases of individual steps

A common requirement for these sensors is that they must be small and light. Most of sensors are passive elements which indicate no need for power. However in the case of the need of power another restriction is added to these sensors. The power consumption of these sensors must be as small as possible.

Processing Unit

The processing unit is considered as the main core for sensor device. Its function is to process data and control all other modules. From many processing devices such as microprocessors, microcontrollers, DSPs, and FPGA, microcontrollers (MCUs) are considered the most common devices used in health care sensors. Microcontrollers represent the integration of many functions in a single small, cheap, and low power chip. In addition, the producing of specific microcontrollers for health care applications makes them more suitable for this task. Microcontrollers used in health care are ultra low power devices which lengthen battery life. They contain some special peripherals in addition to the normal peripherals contained in normal microcontrollers. A precise ADC (analog to digital converter) is an important component which receives the sensor signal. LCD driver can be used to drive LCD display to show information and measurements. DACs, timers, and counters are normal components contained in most microcontrollers. Also integrated flash and RAM memory are important to store data and programs. Interfaces such as I2C, and SPI are used to easily connect microcontrollers to other components. Wireless connection can be established via RF converter contained in some health care microcontrollers.

Sensors storage as a part of the processing unit is usually built in flash and RAM memories. However in many cases they are not enough for storing programs and data. External memories are needed in these cases to compensate small memory capacities. Between many memory technologies, flash memory is considered as an optimum selection for health care sensor devices. Flash is light, compact, energy-efficient, and ever less expensive. There are two kinds of flash namely: linear flash and ATA flash. Linear flash, as the name implies, is laid out and addressed linearly, in blocks. The same address always maps to the same physical block of memory, and the chips and modules contain only memory plus address decoding and buffer circuits. This makes them simple, cheap, and energy-efficient. Linear flash is the obvious choice for nonvolatile memory that's built permanently into an embedded system. ATA flash memory appears as if it was sectors on a hard disk and is accessed via the same register interface used by the original IBM PC/AT's hard disk controller (and, more recently, IDE disk drives).

The main advantages of ATA flash, from the embedded system developer's perspective, are flexibility and interchangeability with hard disks. Flash memory offerings vary widely in capacity, price, speed, and features. This makes designing with flash a non-trivial exercise; the effective embedded system designer must know the full range of available products in order to choose a cost-effective solution. One of the first tradeoffs the designer must consider-as with all embedded systems components-is the inevitable compromise between power and speed. Some flash memories can run at lower voltages (as low as 3V or less), which saves power, but works more slowly. Others run at higher speeds but require five or even 12V.

Communication Unit

Communication module is a vital component in health care sensor to send data to a server to monitor data, store information, or take an action. This server can be mobile phone, laptop, internet, or a hospital. Many technologies are available with different characteristics of power, communication distances, bandwidths, and transfer rates. Bluetooth technology has a range of 100 meters and not necessarily is in sight and the power used by Bluetooth is very less. The nearest competitor of Bluetooth is infrared. Infrared has many additional features but it loses on one point with Bluetooth, as infrared rays cannot pass through the walls and other obstacles. Infrared technology is up in rate of data, Bluetooth has 1MBps whereas infrared has 4 MBps. Infrared is faster than Bluetooth technology. Wi-Fi uses RF waves to exchange data; however Wi-Fi has a larger range than blue tooth. Communicating to a distant server such as hospitals requires another technology. GSM is the best solution for this task.

Power Unit

Many technologies are being focused on how to operated devices with reduced power consumption, but at the same time battery technologies need to catch up with application requirements. There is certainly no shortage of battery- and chemistry-related technologies, ranging from regular lithium-ion batteries to portable rechargeable batteries to fuel cells. Portable rechargeable cell chemistries include Alkaline, Nickel Cadmium (NiCd), Nickel Metal Hydride (NiMH), and Lithium Ion (Li-ion). The Li-ion cells have the highest energy density by both weight and volume. With the appropriate level of safety designed into a Li-ion pack, Li-ion offers the most attractive method of portable battery power.

3. Heath Mentoring Sensors

Before describing some of the sensors applications in the field of health monitoring, it is appropriate to look at the current sensing devices that are used in this field. The following are some of the commercial devices that used separately or part of an application.

Pocket PC

Smart Pocket PCs play an important role in different application. Since they are always carried by a human, it can be used to help people in many different directions. The usage of the Pocket PC in e-health could be essential in applications such as cutting smoking. One of the real example is “ My last Cigarette” project ‎[20]; such project , as shown in Figure 2, displays nicotine level readout , expected cravings readout , daily motivational quote or medical fact, deaths since you quit readout , daily motivational message, and many other features. These readings are extracted through a simple nicotine patch that is connected to the pocket PC holder. It is really impressive how such device can help saving his/her holder life since the holder could be dying for data; the reader is referred to this story to realize the importance of keeping the medical information along with the patient

|[pic] |[pic] |

| Pocket PC |My last Cigarette Software |

|Figure 2: Pocket PC ‎[2] |

Medical Alerts and Recording Devices

Nowadays, human accessories became important medical devices. For instance, the simple bracelets or necklaces, shown in Figure 3, could hold the holder person’s medical information in forms like RFID tag, barcode, or the patient ID is just engraved on the back of it. One of the real applications is the CADEX watch ‎[21]. This watch is used to save the patient critical information such as his/her hospital, ID, and/or insurance information. It can also set different alerts for different medications and times.

[pic]

Figure 3: Simple medical alerts and recording devices ‎[3]

Blood Pressure/Pulse Monitors

Another device, as shown in Figure 4 that can help in early detection of patients’ heart problems is the blood pressure reader or pulse monitor device. Such device is designed with high accuracy and error detection techniques as well as fuzzy logic measurements. In addition, it contains a memory for keeping the measurements history for some time. It is obvious that such device can give an indication to his/her holder by the current situation especially elder people.

[pic]

Figure 4: Blood pressure monitoring device ‎[4]

Wearable Insulin Pumps

Persons with Diabetes are given special care due to the criticality of their cases. A wearable insulin pump device is invited especially for them, e.g. see Figure 5. Such device has a catheter at the end of the insulin pump that is inserted through a needle into the abdominal fat of a person with diabetes. Dosage instructions are entered into the pump's small computer and the right amount of insulin is injected in a controlled manner. It will be more beneficial of such device can transfer the patient’s information to its treatment hospital at the same time to keep his/her record updated.

[pic]

Figure 5: Wearable insulin pump device

AMON - Advanced Telemedical Monitor

AMON is wireless monitoring system that is described at ‎[5]. The system includes a wrist-mounted Monitoring Device (WMD), as shown in Figure 6, with different sensors such as heart rate, heart rhythm, 2-lead ECG, blood pressure, O2 blood saturation, skin perspiration and body temperature sensors. The device is a part of a system that uses these advanced bio-sensors to gather vital information, analyze it automatically using a built-in expert system, and transmit the data to a remote telemedicine centre, for analysis and emergency care, using GSM/UMTS cellular infrastructure.

[pic]

Figure 6: The wrist-mounted Monitoring Device

The “Digital Plaster”

The digital plaster shown in Figure 7 is a device meant to be embedded in ordinary plaster that includes sensors for monitoring health-related metadata such as blood pressure, temperature and glucose levels. The “digital plaster” contains a Sensium silicon chip, powered by a small battery, which sends data via a cell phone or PDA to a central computer database. If the results show any worrisome signs, patients and doctors alike would be notified of the change in the data patterns. This also planned to add a motion sensor to the device so it could additionally serve in the role of “granny monitor” by detecting things like falls or complete inactivity.

[pic]

Figure 7: Digital plaster device ‎[6]

4. Health Care Applications/Prototypes

In this section, we will survey some of the current e-health applications. Some of these applications are very simple to implement. However, many others are complicated since they combine different types of networks and sensing devices.

4.1 Baby Care - Sleep Safe

Numerous studies have found a higher incidence of Sudden Infant Death Syndrome (SIDS) among babies placed on their stomachs to sleep ‎[8]‎[9]. Stomach sleeping puts pressure on a child's jaw, therefore narrowing the airway and hampering breathing. A simple prototype (called SleepSafe) ‎[7] detects the sleeping position of the infant. It alerts the parents when the infant is detected to be lying on its stomach, offering them peace of mind without having to constantly watch their child while it sleeps. This prototype architecture is shown in Figure 8(a).

[pic]

Figure 8: SleepSafe baby monitor for detecting infant sleeping position ‎[7]

The sensor mote attached to the infant’s clothing is a SHIMMER mote ‎[14]. This mote has a 3-axis accelerometer; a single axis is used to sense the infant’s position relative to gravity. Three discrete positions (back, side, and stomach) are measured as anti-parallel, perpendicular, and parallel to the force of gravity. Figure 8(b) illustrates how these positions are measured relative to gravity.

4.2. Baby Care - Baby Glove

As the weight of children decreases, the mortality rate increases ‎[10]. Many of these statistics are due in part to their extreme sensitivity to temperature fluctuations, which must stay within a consistent range of 36◦C to 38◦C. An integrated health monitoring device has been developed to closely monitor vitals, contained within a swaddling baby wrap (called The Baby Glove) ‎[7].

[pic]

Fig. 9 The Baby Glove prototype ‎[7]

The Baby Glove prototype, as seen in Figure 9, includes two sensor network motes, one connected to the swaddling wrap and the other to a base station computer. The first mote, connected to the wrap, is a SHIMMER mote. It monitors the vital information coming from the sensors via an ADC, organizes the measurements into packets and transmits them wirelessly to the second mote, connected to the base station computer, for processing.

4.3. LISTSENse

Hearing impaired people represent a growing percentage of the nation’s population ‎[11]. LISTENse is a prototype that gives the deaf people with the perception ability of critical audible information in their environment (e.g. doorbell, smoke alarm, crying child.) It is comprised of at least two wireless sensor network motes. User carries one mote – the Base Station – on his wrist, belt, etc. and each of the other motes – the Transmitters – is placed close to the sound source that it is to be “heard.” Figure 10(a) shows a simple communication scheme of the LISTENse basic network. Once the measured signal surpasses the reference value, an encrypted activation message is sent to the Base Station. As soon as the Base Station receives the activation message, it extracts the Transmitter address, turns on the vibrator and lights up the corresponding LEDs to warn the user. Figures 10(b) and 10(c) shows the manufactured Base Station and Transmitter prototype‎[7].

[pic]

Figure 10 LISTENse prototype ‎[7]

4.4. CodeBlue

CodeBlue is a working prototype of a wireless, web-enabled, health monitoring system. CodeBlue consists of three functional components: a wearable unit, a base station, and a web server. Figure 11 illustrates the CodeBlue software platform.

[pic]

Figure 11: CodeBlue architecture for emergency response‎[15].

The wearable unit collects the patient’s physiological parameters via sensors attached to the patient’s body. It then transmits this data wirelessly to a base station where it is analyzed and stored. The base station automatically contacts the patient’s nurse and/or paramedics when measured value surpass abnormal. The web server component allows medical professionals to remotely access the patient’s physiological data over the Internet, and provide feedback to the patient by phone or text-messaging ‎[15]‎[16]‎[17].

4.5. Smart Home Care

Smart homecare may assist residents by providing memory enhancement, medical data lookup, and emergency communication. The data collected from automatic monitoring via wireless sensor network can be stored and integrated into a health record of each patient, which helps physicians make more informed diagnoses. Also, quickly notifying doctors of any changes in vital signals may save human lives ‎[18]‎[19].

[pic]

Figure 12. Smart home and personalized health monitoring architecture

Figure 12 illustrates the architecture of the smart home care system in ‎[13]‎[13]. It consists of two parts namely the smart phone application and the healthcare centre server. The smart phone communicates with the home server via WiFi or cable. The home server runs software to control the webcams and controls the information exchange with the health care server via an ADSL connection.

5. Health Care and Expert Systems

Expert systems play a great role in health care field. It depends on knowledge representation and inference engines. Two different approaches to represent knowledge in expert systems have been followed. In the first approach, diseases and clinical, physiological, or pathophysiological states are each characterized by a stored pattern of anticipated findings to be matched with the clinical findings observed in the patient. These patterns, referred to as nodes or frames, are interconnected by links representing various relations and dependencies. Such networks of nodes and frames are excellent to represent hierarchical knowledge structures. Newer programs (second-generation programs) are employing networks where the use of causal pathophysiological knowledge is emphasized. A typical feature of these programs is a taxonomic structuring of the nodes and links that allows diagnostic and other types of reasoning to proceed at various levels of detail.

The second approach is based on a representation of the knowledge in the form of rules (production rules, IF-THEN rules), which are applied in an orderly manner to produce the solution of a diagnostic problem.

A production rule consists of a set of preconditions, referred to as the premise, and an action part. If the premise is true, the conclusion in the action part is justified. A prominent example of a system based on the production rule formalism is the MYCIN system ‎[22], which provides consultation about infectious disease, diagnosis, and choice of therapy.

Each conclusion includes a certainty factor (CF) ranging from 1 (complete belief) via 0 (nothing known) to -1 (complete disbelief), and each assertation is associated with a CF. If the CF on a premise is positive above a certain threshold (0.2), the corresponding conclusion is drawn with a certainty that is equal to the premise’s CF times the conclusion’s own CF. Evidence confirming a hypothesis is collected separately from that which disconfirms it, and the truth of a hypothesis is the sum of the evidence for and against the same hypothesis. MYCIN knows that its goal is to undertake a number of tasks such as (e.g.) to determine if there is a substantial infection in the patient.

To accomplish a goal, MYCIN evaluates all rules relevant for this goal. This creates a need to evaluate the premises of these rules, and these then become new subgoals, which are treated in the same way-i.e., the rules relevant for these subgoals are evaluated. When no rules are found that apply to a subgoal, the user is asked to supply the needed information. This way of searching for a solution is referred to as “backwards chaining”. A “forward chaining” or “data driven” approach is used when the patient’s data are entered without guidance by the computer. Those rules whose premises match the data are then applied, and new rules that use the conclusions in their premise conditions are subsequently applied, etc. Instead of using one of the two strategies, it is also possible to combine them into a mixed strategy.

1. E-health Expert Systems Classification

Table 1 lists some examples of expert systems. The majority of systems are diagnostic systems that output a diagnosis and, possibly, suggestions for treatment. In addition to diagnoses, some systems such as the ABEL ‎[23] make predictions, e.g., about possible effects of different therapeutic interventions. Of particular relevance for the clinical pathologist is the planning involved in the patient workup, because this has a bearing on the efficient utilization of the laboratory services.

PHEO-ArFENDING is an expert system that assists the physician in this planning ‎[30]. It is designed to critique a physician’s workup of a patient with suspected pheochromocytoma. In contrast to more traditional systems, this system first asks the physician to describe his (or her) patient and to outline the approach planned. The system then critiques that plan to help the physician make the workup as rational and efficient as possible. A recommended sequence of work-up is built into the system’s knowledge base, which is organized as expressive frames associated with each test or procedure. Each frame contains a list of comments that may be output in discussing the use of the test or procedure. Each comment has an associated condition that indicates when it is to be output as part of the critique. For instance, if a CT scan is ordered without prior screening tests, a comment in the CT scan frame suggests that these should be ordered first and that a CT scan is not indicated. The various comments generated by activating the various frames during a consultation are combined into a smooth narrative text, which is then output as a final critique.

Table 1: Some Expert Systems Emphasizing the Use of Laboratory Data

|System reference |Domain |Use |Reference no. |

|ABEL |Acid-base, electrolytes |Diagnosis, prediction |‎[23] |

|ANEMIA |Anemia |Diagnosis |‎[25] |

|Consult-I |Anemia |Diagnosis |‎[26] |

|PAThFINDER |Lymph-node histopathology |Diagnosis |‎[27] |

|EMYCIN |Leukemia |Diagnosis |‎[28] |

|RED |Erythrocyte antibodies |Diagnosis |‎[29] |

|PHEOAUENDING |Pheochromo-cytoma |Critique of patient work-up |‎[30] |

|EXPERT |Serum proteins |Diagnosis |‎[31] |

|PRO.M.O. |Lipoprotein metabolism |Diagnosis |‎[32] |

|LIThOS |x-ray anal. of renalstones |Diagnosis (of stone content) |‎[33] |

|EXPERT |Outpatient testing |Diagnosis, planning test requests |‎[34] |

|LIVER |100 diseases |Diagnosis |‎[35] |

|SMR |Multiple |Diagnosis, interpretive comments |‎[36] |

Van Lente et al. ‎[34] used the EXPERT system ‎[24] for sequential laboratory testing and interpretation in an outpatient setting. Abnormalities in the initial test profile initiate a sequential laboratory testing of the specimen already collected. The incremental charge to a patient evaluated by this sequential testing program was only slightly more than the phlebotomy charge that would be incurred if one additional specimen were required for the physician to investigate an abnormality independently. Some expert systems have been limited to instruments.

Weiss and Kulikowski used the EXPERT building tool ‎[24] to construct a system that interprets serum electrophoresis patterns ‎[31]. This was later incorporated in a densitometer. Wulkan and Leijnse ‎[33] reported a system in which a PC connected to an x-ray diffraction analyzer system preprocesses the diffraction data on-line and stores selected data in a LISP format. These data are subsequently interpreted by a rule-based expert system LIThOS, which reports the components and relative content of renal stones. The system was developed because of a shortage of the specialists needed to interpret diffractograms.

2. Online Community with Ask the Expert System

There is an increasing number of online health communities on the Internet for people with different health conditions ‎[37]‎[38]‎[38]. Studies have shown that patients who interact online with people who have similar health problems benefit from this ‎[39]. Some of the health-communities are for people who face problems caused by their life-style, who want to change behavior. Empathy and advice are communicated, and to some extent are also questions concerning ideas and beliefs raised ‎[40]‎[41]. Examples of health-communities for lifestyle problems are communities for people, who try to quit smoking, lose weight or give up drugs or alcohol.

In online communities people can think together, ask questions, guide each other, and share ideas and insights ‎[42]. Through interaction with others, new ideas and knowledge can be developed ‎[43], and we are able to understand our situation better ‎[42].

Learning is always a challenge. Especially challenging is learning how to change behavior, to unlearn a bad habit and to develop a healthier way of living.

Another type of online system used for gaining new knowledge and to get help in developing new practices is the so-called Ask the Expert systems. They facilitate learning through giving recommendations, advice, etc. from a medical expert ‎[44], and they offer a new type of continuous relationships between patients and medical experts ‎[45]. During recent years, studies have been conducted of the usage of different knowledge management tools, such as online communities for different purposes.

3. Health Care Management

The goal of health care management is to ensure that high quality patient care is delivered in the most affordable way possible. Health care management attempts to achieve this goal by stressing the importance of preventive care and trying to reduce the over-utilization of surgery and hospital admissions. Another important component of this care is medical case management, in which potentially catastrophic or chronic cases are given special attention. Health care management services are delivered by a team of medical review specialists, case managers, and physicians. The medical review specialists and case managers are registered nurses by training. The health care management providers combine general clinical expertise with the specialized skills needed to effectively deliver affordable health care.

To provide support for delivering health care management services, several expert systems (e.g. INFER, PsychINFER, Procedure Necessity, and Alternatives to Non-Surgical Admissions (ANSA) ‎[46] have been developed. INFER and PsychINFER help a medical review specialist to identify potentially catastrophic or chronic cases so that they may be aggressively managed. Procedure Necessity helps to decide whether a surgical or diagnostic procedure is necessary on a case-by-case basis. ANSA gives advice about whether a hospital admission is appropriate in cases where a patient is being admitted to a hospital for a reason other than to have surgery.

5. Open Problems

There are many open problems in e-health; these problems share some of the ad hoc and sensor networks problems. A sublist of these problems is introduced as follows:

1- Patients’ Privacy: this a social issue that is heavily expressed in Europe and United State. Patients are always worried about their health information.

2- Security: As in any field, information security is essential task. However, in case of health information, patients’ lives depend on such information. Altering or playing with patients information may lead to a disaster.

3- Sensors Integration: The current e-health applications use many of the sensing devices that are made available by different vendors. Some of these sensors are not standardized. Therefore, integrating such sensors in one application still an open problem.

4- Data Mining: Patients’ information is expected to be huge which leads to data mining problem. This problem needs to be investigated. Smart storage is required for retrieval purposes.

6. Conclusion

In this paper, we surveyed the e-health technology, application, as well intelligent and expert health care applications. As can be seen, the field of e-health still young and needs a lot of research effort. This paper is intended to guide the reader to the current as well as weakness or open problems in this field. Our future work focuses on benefiting from the sensor networks and its intelligence in collecting the information towards the e-health and data manipulation.

References

1] Emil Jovanov1, Aleksandar Milenkovic1, Chris Otto1 and Piet C de Groen2 A wireless body area network of intelligent motion sensors for computer assisted physical rehabilitation Journal of NeuroEngineering and Rehabilitation 2005

2] Pocket PC,

3] Simple medical alerts and recording devices,

4] Blood pressure monitoring device,

5] Anliker, U.   Ward, J.A.   Lukowicz, P.   Troster, G.   Dolveck, F.   Baer, M.   Keita, F.   Schenker, E.B.   Catarsi, F.   Coluccini, L.   Belardinelli, A.   Shklarski, D.   Alon, M.   Hirt, E.   Schmid, R.   Vuskovic, M.  Wearable Lab., Swiss Fed. Inst. of Technol., Zurich, Switzerland; “AMON: a wearable multiparameter medical monitoring and alert system,” IEEE Transactions on Information Technology in Biomedicine, Volume: 8, Issue: 4, pp 415-427, 2004.

6] Digital plaster device,



7] Baker, C., Armijo, K., Belka, S., Benhabib, M., et al. “Wireless Sensor Networks for Home Health Care”, 21st International Conference on Advanced Information Networking and Applications Workshops, 2007. AINAW ’07 (21-23 May 2007): Volume 2, 832-837.

8] American SIDS Institute Website, , 2006.

9] Kids Health Website, , 2006.

10] March of Dimes, , 2006.

11] Center for Disease Control, , 2006.

12] Tarek R. Sheltami *, Ashraf S. Mahmoud, and Marwan H. Abu-Amara, “An Ad Hoc Wireless Sensor Network for Telemedicine Applications”, The Arabian Journal for Science and Engineering, Volume 32, Number 1B.

13] Peter Leijdekkers, Valerie Gay, Elaine Lawrence, "Smart Homecare System for Health Tele-monitoring," pp.3-7, First International Conference on the Digital Society (ICDS'07), 2007

14] , 2006

15] Division of Engineering and Applied Sciences, Harvard University, “CodeBlue: Wireless Sensor Networks for Medical Care”,



16] Tia Gao, Christopher Pesto, Leo Selavo, Yin Chen, JeongGil Ko, JongHyun Lim, Andreas Terzis, Andrew Watt, James Jeng, Bor-rong Chen, Konrad Lorincz, Matt Welsh, “Wireless Medical Sensor Networks in Emergency Response: Implementation and Pilot Results”, IEEE International Conference on Technologies for Homeland Security, May, 2008

17] David Malan, Thaddeus Fulford-Jones, Matt Welsh, Steve Moulton, “CodeBlue: An Ad Hoc Sensor Network Infrastructure for Emergency Medical Care”, MobiSys 2004 Workshop on Applications of Mobile Embedded Systems (WAMES 2004), June, 2004

18] Jakkula, Vikramaditya R.   Cook, Diane J.   Jain, Gaurav, “Prediction Models for a Smart Home Based Health Care System”, 21st International Conference on Advanced Information Networking and Applications Workshops, 2007, AINAW '07, Washington State University, USA; Volume 2,  761- 765

19] Helal, A.   Abdulrazak, “Toward a Scalable Home-Care Delivery for Frail Elders and People with Special Needs”, ICORR 2007. IEEE 10th International Conference on Rehabilitation Robotics, June 2007, 994-998

20]

21]

22] Shortliffe EH. Computer-based medical consultations: MYCIN. New York: Elsevier, 1976.

23] Patil RB, Szolovits P, Schwartz WE. Modelling knowledge of the patient in acid-base and electrolyte disorders. In: Szolovita P, ed. Artificial intelligence in medicine. Boulder, Colorado: Westview Press, 1982:191-226.

24] Weiss SM, Kulikowski CA. EXPERT: a system for developing consultation models. In: Proc Sixth Int’l Joint Conf on Artificial Intelligence. Tokyo: Information Processing Soc. of Jpn. 1979;942- 7.

25] Quaglini 5, Stefanelli M. ANEMIA: an expert consultation system. Comp Biomed Res 1986;19:13-27.

26] Blomberg DJ, Ladley JL, Fattu JM, Patrick EA. The use of an expert system in the clinical laboratory as an aid in the diagnosis of anaemia. Am J Clin Pathol 1987;87:608-13.

27] Horvitz EJ, Heckerman DE, Nathwani BN, et al. Diagnostic strategies in the hypothesis-directed PATHFINDER system. In: Proc First Conf Artificial Intelligence Appl. Silver Spring, MD: Institute of Electrical and Electronics Engineers Computer Society, 1984;630-6.

28] Fox J, Myers CD, Greaves MF, et al. Knowledge acquisition for expert systems: experience in leukemia diagnosis. Methods Inf Med 1985; 24:65-72.

29] Smith JW, Svirbely JR, Evans CA, et al. RED: a red-cell antibody identification expert module. J Med Syst 1985; 9:121-38.

30] Miller PL, Blumenfrucht SJ, Black HR. An expert system which critiques patient workup: modeling conflicting expertise. Comp Biomed Res 1984;17:554-69.

31] Weiss SM, Kulikowski CA, Galen RB. Representing expertise in a computer program: the serum protein diagnostic program. J Clin Lab Automation 1983;3:383-7.

32] Trendelenburg C. Routine applications of the expert system PRO.M.D. In: KerkhofPLM, van Dieijen-Visser MP, eds. Laboratory data and patient care. New York and London: Plenum Press, 1988.

33] Wulkan RW, Leijnse B. Experience with expert systems in clinical chemistry. Ibid.

34] Van Lente F, Castellani W, Chou D, Matzen RN, Galen RB. Application of the EXPERT consultation system to accelerated laboratory testing and interpretation. Chin Chem 1986;32:1719-25.

35] Chang E, McNeely M, Gamble K. Strategies for choosing the next test in an expert system. In: Proc Amer Assoc Med Syst and Informatics Congress 1984. Bethesda, MD: American Association of Medical Systems and Informatics, 1984:198-202.

36] Wiener FM, Groth T. A system for simulating medical reasoning (SMR) providing expertise in clinical computer applications. Automedica 1987;8:141-9.

37] J. Preece, Online Communities – Designing Usability, Supporting Sociability, Wiley & Sons, 2000.

38] S. Fox, D. Fallows, “Internet health resources”, Pew Internet & American Life Project, July 16, 2003.

39] J. B. Walther, S. Pingree, R. P. Hawkins, D. B. Buller, “Attributes of interactive online health information systems”, Journal of Medical Internet Research, Vol. 7, Issue 3, 2005.

40] Å. Smedberg, “Learning through online communities – a study of health care sites in Europe”, proceedings of e-Challenges Conference, Vienna, Austria, 27-29 October 2004 (eAdoption and the Knowledge Economy: Issues, Applications, Case Studies, vol.1, edited by P. Cunningham and M. Cunningham, IOS Press, 2004, pp. 133-1339).

41] Å. Smedberg, “Double-loop learning conversations in an online community on overweight”, proceedings of IADIS International Conference on Cognition and Exploratory Learning in Digital Age, CELDA 2005, Porto, Portugal, 14-16 December 2005.

42] R. McDermott, “Knowing in community: 10 critical success factors in building communities of practice, IHRIM Journal, March 2000.

43] K. Starkey, S. Tempest, A. McKinlay, How Organizations Learn, 2nd edition, Thomson Learning, 2004.

44] R. Bromme, R. Jucks, T. Wagner, “How to refer to ‘diabetes’? Language in online health advice”, Applied Cognitive Psychology, Vol. 19, Issue 5, 2005, pp. 569-586.

45] J. Marco et al., ”Advice from a medical expert through the Internet on queries about AIDS and hepatitis: analysis of a pilot experiment”, PLOS Medicine, Public Library of Science, Vol. 3, Issue 7, July 2006.

46] Glenn J. Fala, Kathryn T, Clayton, and Diane M. Masciantonio, “Applying expert systems to health care management “, ACM O-89791-658-1 95 0002 3.50, 1995.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download