Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Distributed Mobile Services and Interfaces for People Suffering from Cognitive Deficits Sylvain Giroux, Hélène Pigot, Jean-Pierre Savary Jean-François Moreau Département d’informatique Division R&D CRD Faculté des sciences France Telecom, Université de Sherbrooke 32 rue de la République 2500, boul. Université, 92130 Issy les Moulineaux Sherbrooke, Qc France Canada, J1K 2R1 Phone : (33) 1 45 29 42 47 Fax : (33) 1 45 29 62 90 Phone : (1)-819-821-8000 Ext 2027 Fax : (1)-819-821-8200 e-mail: [email protected] [email protected] [email protected] e-mail: [email protected] Distributed Mobile Services and Interfaces for People Suffering from Cognitive Deficits KEYWORDS Mobile Services, Cognitive Assistance, Smart Home, Cognitive Deficits, Reminder, Remote Monitoring, Pervasive Computing, Sensors, Batch Mode, Direct Mode. INTRODUCTION The number of people suffering from cognitive impairments (Alzheimer disease, head injury, schizophrenia...) is growing continually. For one, the ageing of the society plays an important role in this trend. The always increasing needs for resources generates a huge stress on human and economical actors. Thus in Canada and United States, as in most western countries, social and medical policies try to keep at home people suffering from cognitive deficits [Hareven, 2001]. Progress in cognitive rehabilitation also augments the number of semi-autonomous people that would be able to stay at home if light assistance is provided. Of course this results in higher expectations and even more demand on resources. But most of the time, families have to take responsibility for care without access to appropriate resources. Too often, this situation then turns to an exhausting weight. Therefore natural and professional caregivers urge for help. Recent advances in mobile technology can provide affordable solutions to lessen the burden and anxiety put on caregivers and to collect reliable information. For many years, numerous devices have been designed to counterbalance physical and sensory deficits. Nowadays progress in technology set high hopes for cognitive orthotics that address cognitive decline [Pollack, 2004, Lange, 2002]. Thanks to their small size and their versatility, mobile devices can offer a personalized assistance anytime anywhere. Mobile devices then become portable cognitive prosthesis, for instance acting as pervasive system to remind people Activities of Daily Living (ADL) to perform when needed. They can also foster sense of security by keeping people and caregivers directly in touch. In its simplest form a direct call button may be used to request immediate assistance. Geo referenced data may also be used to detect crisis of schizophrenia or when a patient suffering from Alzheimer disease is lost [Médical Intelligence, 2005]. Mobile devices can also provide non intrusive remote supervision by caregivers. Besides mobile devices can gather ecological data compulsory to adapt or finetune diagnosis and treatments. Nevertheless the design of devices and mobile services must put a very careful attention to user interfaces used to deliver information. Population suffering from cognitive deficits has often severe limitations and constraints. This chapter discusses features and implementation of mobile services for cognitive assistance and remote supervision. Mobile devices provide practical solutions to the issues presented above. First we review benefits and limitations of current reminder systems that can assist patients and/or caregivers. Then we present the needs of the population regarding the assistance necessary to stay safely at home. The multiple mobile services, designed especially for that population, are also described. We show how supervision can safely foster autonomy. Next we describe cognitive assistance to patients and remote supervision by professionals or relatives, and then we go further and gather ecological data to foster better treatments. Finally outdoors, safety and security of the patients is relying on geo-localization features. Implementation specific details are also sketched. MOBILE PROSTHETIC SYSTEMS The management of ADL is a central issue for people suffering from cognitive deficits. In the process of rehabilitation, occupational therapists provide a patient with a paper agenda as a tool he has to master to manage his life and autonomy. A lot of research projects and commercial applications also targeted electronic adaptation of specific functionalities of these agendas. In this section, we review the advantages and limitations of current reminder and / or agenda systems. Some systems are device specific and are explicitly designed to be used inside the patient home and then could not help outside [Visions, 2005], [Helal, & al., 2003]. Another category of systems is designed specifically for mobile devices in such a way that their use is not restricted to a specific location. These applications are usually running on PDAs or smart phones to either assist for needs specifically related to mobility [Patterson et al., 2004] or as general reminder/agenda systems or electronic organizers [Szymkowiak, 2005] [Gorman, 2003] [Haberman, 2005] [Neuropage]. They are usually provided with acoustic alert or remote communication. Activity Compass project intends to provide compensatory aid for outdoor courses for patient suffering from Alzheimer disease [Patterson & al., 2004]. It consists basically of a PDA equipped with a GPS. Its role is to guide patient towards their destination by a mean of an arrow indicating the direction to take. Neuropage is a reminder system that uses radio technology to send reminders of things to do [Neuropage]. When the message arrives, the pager beeps (or vibrates), one button is pressed and the message can be read from the screen. Messages can be regular events or single-time message. Messages are added or removed by contacting Neuropage office by phone, email, letter, or fax. The ISAAC system acts as a cognitive prosthesis used by people experiencing dysfunctions in autonomy due to different kinds of brain injury [Gorman & al., 2003]. It provides a checklist reminder to execute safely the ADL. Its hierarchical structure permits to navigate easily throughout the different pieces of advice displayed. Personalization and adaptation of pieces of advice enable ISAAC to evolve according to the patient cognitive evolution. Providing a reminder helps the cognitively impaired patient to increase his autonomy. But the patient himself and his caregiver express needs for safety. Both want to keep in touch and be confident that activities will be performed in time. So a mean of communication must be added to the current reminders. In the next section we review the cognitive deficits encountered and the needs that ensue. NEEDS OF COGNITIVELY IMPAIRED PEOPLE Cognitive impairments encountered in schizophrenia, head trauma and during the early stages of the Alzheimer disease provoke similar losses in autonomy that justify exploring a common approach to support, assistance and remediation by technology [Pigot, et al., 2005]. Cognitively impaired people present lacks of initiation. For instance, they remain for long periods without undertaking actions. They all call for frequent reminders to remember what to do and sometimes how to do. Alzheimer disease strikes elders and one’s situation evolves from relative autonomy to an unrelenting dependence while trauma injury and schizophrenia appear most often during adulthood and one’s situation evolves from dependence to relative autonomy. Unless severely affected, cognitively impaired people are usually quite autonomous in performing basic ADL such as eating, dressing, and washing themselves. But they at times forget to do them and then need continuous recalls in order to initiate ADLs. Taking medication is a critical issue, as it requires a good short-term memory in order to fulfill the prescription correctly: which pills, how many and when. Without a reliable short-term memory, it can be forgotten or taken twice. At home oblivion or bad discernment in the use of domestic appliances often cause bath and sink flooding, fire, burn, cut... Beside these hazardous situations, risk of malnutrition, bad hygiene and isolation are also real. Cognitive impairments experienced at home have also manifest effects outside. Moreover cognitively impaired people exhibit more difficulties in an unusual context. Oblivion leads to problems in finding one’s way to a destination, in remembering the shopping list or even the reason of the trip. Loss of autonomy requires assistance. Caregivers lavish attention to the patients days and nights long. Frequent recalls and safety are two factors especially important in the overall burden coming from constant supervision and assistance. Therefore an electronic companion could alleviate the “threatening” recalls and improve the relationship. Safety prevents a caregiver from letting the patient alone. MOBILE AND PERVASIVE COMPUTING Although existing services are invaluable for patients and caregivers, overall they suffer from some drawbacks. These services are not integrated in a common portal. They must run on a specific patient device. They are either closely linked to a given location or, if they are location independent, they can not benefit from the casual presence of other devices in the environment, for instance sensors on the kitchen cupboards. In next sections, we present the first steps of an on-going research project at DOMUS Laboratory. Mobile computing is coupled with pervasive computing [Weiser, 1991]. Mobile computing will enable patients and caregivers to use services wherever they are. Pervasive computing will allow them to access services whatever the device is and to benefit from existing devices, sensors and effectors present in their current environment. This approach then promotes a tighter integration of devices and services to the environment when possible while preserving mobility and independence of the patients. Currently a common portal and many services were implemented and are ready to be evaluated in situ. Through this portal a patient gets access to services with a simplified adapted user interface. Caregivers have also access to their counterpart of services but the interface is more complex. For the moment, devices used are PDAs, but we intend to port them on e-mate, a platform that renders deployment of services independent of devices they are deployed on [Giroux & al., 2003]. So phones, laptops, TVs could be used to access services. ASSISTANCE AND SUPERVISION SERVICES A mobile device is designed to offer several services to enhance autonomy, security and communication to cognitively impaired people and their caregivers. PDAs are linked through a server. Some are dedicated to patients. Others run applications presenting the caregivers point of view on information. Patient services fill three functions: a simplified reminder, a prompter to request immediate assistance, a service to gather ecological information and a location based health information service [Boulos & Maged, 2003] to help patient in case of crisis. On the other side, three services are available for the caregiver: monitoring patients’ ADLs, informing system and colleagues of an intervention, and planning patient’ ADLs. Services also enable patients and caregivers to communicate. The client component running on a patient’s PDA is kept very simple. The patient benefits on an auto login procedure in order to reduce the cognitive load. An Enhanced Agenda as a Mean for Cognitive Assistance A basic paper agenda is designed to help remembering appointments, to fill a timetable and to prevent time conflicts. Our aim here is quite different. The agenda acts mainly as a cognitive prosthesis. Typically the home of a patient suffering from head injury is covered with numerous notices, for instance written on a post-it. Each one indicates something to do. On the contrary, the electronic diary can show solely the activities often forgotten, and moreover just when it is necessary. By default, our prototype displays on the PDA single window the next three activities to perform in a three hours time bracket (Figure 1). Depending on the patient’s cognitive abilities the number of activities displayed and the time slot can be easily reduced or increased. When an activity is performed the patient clicks on it to indicate its completion to the caregivers. The activities monitored on the agenda are either events occurring just one time, as appointments, or recurring events as ADLs often forgotten. Figure 1. Patient agenda. Keeping in Touch If help is necessary, the patient clicks on the button “Help” at the bottom of the screen to warn caregivers. This button is always visible and active. This remote supervision ensures permanent contact between them. It provides a sense of safety for both patients and caregivers. Caregivers feel insecure to let patients alone without the opportunity to be reached. For patients, the phone is often too complicated to use. Carrying everywhere a PDA offers permanent assistance at a single click. Monitoring ADL Previous sections described the patient side of the system. In this section and the next two, we depict the caregiver view: how ADLs are monitored, how cooperation can be foster between caregivers and how to manage a patient list of ADLs. The role of the patient PDA is to foster his autonomy. From time to time, he does not perform given activities in time which may compromise his ability to stay alone. Equilibrium must be maintained between a patient intimacy and his safety. For instance, forgetting one day to take a shower does not justify an immediate intervention. However after several days without showers, caregivers urge him to do it. On the other side, forgetting once to take medication could be harmful for the patient’s health. Without a tool that helps to monitor ADLs, a caregiver surveys constantly the patients asking for the ADLs completion. (a) (b) Figure 2: Caregiver’s ADL monitoring. a) Supervised patients. b) One patient ADLs The current prototype supports remote supervision of ADLs to perform (Figure 2). Caregivers then know when an ADL is completed without bothering residents. On his mobile device, a caregiver can monitor every patient under his responsibility. On his patients list, colored icons indicate the status of current ADLs for each patient: green if all is OK, yellow if the patient is near to forgot to perform an ADL and red if time planned for the ADL is over (Figure 2a). For each patient, a caregiver can know: which activities are under monitoring (Figure 2.b) and the time of completion of a given ADL. The same color code prevails. Coordination between Caregivers Since many caregivers can be in charge of a patient, coordination is important. Several caregivers can take care of many patients. When a patient does not perform a task in time, all responsible caregivers are informed. It is then essential to coordinate their actions, for instance in order to avoid all of them going to see the patient. As soon as a caregiver notifies his intervention to the system, this information is sent to all the mobile devices owned by the other caregivers. Caregivers can also exchange responsibility of patients, for instance when a working shift occurs. Implementation section explains in more details how the information is updated on each PDA. Planning ADL In addition to ADL monitoring, caregivers must have also simple means to update and change rapidly the ADLs and events that have to be monitored. Like a usual organizer, the caregiver mobile device enables to register an activity occasionally or on a periodic basis ranging from hours to months (Figure 3a). The caregiver has to specify duration and completion time and for how long the activity has to be monitored (Figure 3b). This last information is particularly suitable during the rehabilitation period or to update a medication prescription that is regularly changed by the physician. (a) (b) Figure 3: Planning activity. a) Activity period basis. b) Time activity information. Ecological Data: Gathering Information for Better Treatments Collecting ecological information may be of tremendous value for delivering better treatments. For instance to control the schizophrenic symptoms, a physician prescribes neuroleptics which induce severe side effects [Shriqui and Nasrallah, 1995]. The physician has then to adjust medication during years to limit side effects and to avoid the recurrence of hallucinations. He needs valid information about the patient feelings. On his mobile device, the patient can notify symptoms (or other information) as soon as he feels them (Figure 4). An intensity scale permits him to estimate the level of severity. The PDA is also equipped with a GPS to register automatically date, time and position. These georeferenced data are sent to the server on a batch mode. Data can be analyzed later to improve diagnosis and treatments. Figure 4: Patient symptoms IMPLEMENTATION The basis of the implementation is a client-server architecture (Figure 5). Patients and supervisors use iPAQs to connect to the server through wireless networks. The system is available both indoors and outdoors. Indoors, the wireless connections to the server use IEEE 802.11b,g. Outdoors, the wireless connection is made through a Sierra Aircard. Each PDA connects to the server through TCP sockets. The server allocates a different thread to each connection. If the connection is lost, the mobile device tries to restore it. The PDAs and the server exchange serialized objects through sockets. Outdoors, a GPS gives the patient location. To ensure rapid communication the traffic is kept low. When a patient confirms an ADL has been done or when a caregiver let know about his intervention, a message is sent to the server. The server first updates the database, and then sends a message to each PDA to inform a change happens. At that moment, some PDAs may want to display the information under change. They request all the information needed to update their screen displayed. Figure 5: Connection infrastructure. Different types of communication are used between the PDAs and the server. Inside, when the connection is established the transmission is in direct mode. Outside or when the connection is out, the communication is in batch mode. Messages are saved in a queue waiting till a connection becomes available. Finally a configuration file is used to personalize the application: auto-login, font, font size, message strings, icons, and number of activities displayed. The implementation is in standard Java (J2SE) for the server part. PersonalJava is used for the client part on PDAs. On the PDAs, user interfaces were implemented with SWT. On the PDAs, the virtual machine was J9 for PocketPC. The system was tested on HP iPAQ 4155 and HP iPAQ 1945 running Microsoft PocketPC OS. DISCUSSION Specific needs are addressed by our system for cognitively impaired people in order to foster their autonomy. Functions of an electronic agenda are redesigned to limit cognitive load, especially the burden put on memory. Presently only the patient name is displayed. We are evaluating the relevance of adding other information such as the patient address and caregiver phone number. It may be helpful when a patient is lost but it could raise the cognitive load by requiring one more screen. The help function is germane to a panic button provided by various companies. A panic button is generally linked to a remote controller that intervenes when the button is pressed. The use of such panic systems remains generally difficult especially for elders suffering Alzheimer disease. First, we hope that providing on the PDA a help button they use frequently is more likely to be used in case of emergency. Secondly, making explicit the help button on the PDA by an icon or by writing the function is a guarantee they could understand its use. Providing geo-referenced data is another critical issue. Knowing a patient position may be worthful and even vital in many situations and for a range of purposes: to send directly somebody at the patient location if he is in crisis or in danger; to forbid specific zones; to detect crisis states by analyzing paths and detecting pathologic patterns of movement of patients suffering from schizophrenia; to predict the goal of a patient combining path history and analysis. When a patient is outdoors, a GPS connected to his PDA get his position. This position is sent at predefined time interval to the server. Path analysis can then be performed. Currently client and server components are under development to specify and manage forbidden zones. Mobile services are actually designed to offer the activities monitoring by professional caregivers. Families need too a similar system to be informed of difficulties. They feel insecure to let alone the patient. They could exploit geo-referenced data in the timetable information to know if it is conform to the one expected. For instance a patient suffering from schizophrenia shows disorganization symptoms before experiencing a new crisis. Sending no information about the activity performed is a premonitory sign. Applying it to the Alzheimer disease may help to detect falls or other health problems. CONCLUSION In this chapter, we presented a client-server based mobile system for patients suffering from cognitive deficits and their caregivers. Users will use mobile devices. They help especially to remember the activities a patient have to perform. We adapted traditional agenda to decrease the cognitive prerequisites and to provide a permanent communication between a patient and his caregiver. All along the development process attention was paid on the real needs expressed by patients and caregivers. A sociological questionnaire and study had established previously the needs among patients suffering Alzheimer disease, head injury and schizophrenia [Pigot, & al, 2005]. Periodic meetings with caregivers have ensured conformity of the design of services to the needs expressed. Finally a clinical validation is under process among people suffering from schizophrenia. Future works will integrate mobile services to a helpful environment designed for people suffering from cognitive impairments in the spirit of pervasive computing. For instance instead of asking a patient to notify when an activity is completed, the environment will be able to recognize the activities performed by means of sensors [Pigot, 2004]. The home will adapt itself to alleviate autonomy, remembering when necessary what needs to be done and how to do it. Such smart homes are an answer to the growing demand of cognitively impaired people to stay safely at home. INDEX Mobile Services : Mobile services refer to the software applications available on portable devices such as PDA or smart phone. Cognitive Assistance : It is the assistance provided by the environment to compensate the cognitive deficits. Cognitive Deficits : They are the deficits encountered by a person following a brain lesion. The memory losses and the lacks of planning, initiation and attention are examples of cognitive deficits. Smart Home : The smart home are the home equipped with sensors and effectors that could react according to the inhabitant actions. The smart home could warn persons outside by the remote monitoring or give advices to the inhabitant. Remote Monitoring : see smart home. Reminder : a reminder is a prosthestics aid which displays the common agenda functions and information on memories or advices to be followed. Pervasive Computing : Pervasive computing is the next generation computing environments with information and communication technology everywhere, for everyone, at all times. Sensors : the sensors are electrical devices that gather information on the state of the environment and on the localization and the activities performed by the inhabitant. For instance some sensors detect the presence of the inhabited or which door is open. Direct Mode : The direct mode indicates the transmission mode where the data are sent to the receptor immediatly after the transmittor command. Batch Mode : The batch mode indicates the transmission mode where the data are sent to the receptor a moment after the transmittor command. REFERENCES Boulos, K., & Maged N. (2003) Location-based health information services: a new paradigm in personalised information delivery, International Journal of Health Geographics 2003, 2:2, http://www.ij-healthgeographics.com/content/2/1/2 Giroux, S., Carboni, D., Paddeu, G., Piras, A., & Sanna, S. (2003) Delivery of services on any device : from Java code to user interface. 10th International Conference on Human - Computer Interaction 2003, June 22-27, 2003, Crète, Grèce. in In Constantine Stephanidis (ed.) and Julie Jacko (ed.) HUMAN-COMPUTER INTERACTION, vol. 1-2, Laurence Erlbaum Associates. Gorman P., Dayle R., Hood, C.A., & Rumrell L. (2003) Effectiveness of the ISAAC cognitive prosthetic system for improving rehabilitation outcomes with neurofunctional impairment. NeuroRehabilitation. 18: 57-67 Haberman V., Jones, M. & Mueller, J. (2005) Mobile Technology, Compensatory Aids, and Usability Evaluations. Pervasive Computing. 4 (2) 82-83. Hareven, T.K. (2001). Historical Perspectives on Aging and Family Relations, in R.H. Bintock & L.K. George, (Eds), Handbook of Aging and the Social Sciences, New York, Academic press, 141-159. Helal, S. & al. (2003) Smart Phone Based Cognitive Assistant. UbiHealth 2003: The 2nd International Workshop on Ubiquitous Computing for Pervasive Healthcare Applications, Seattle, Washington, October 12, 2003. Lange, M. L (2002). Technology and occupation : contemporary viewpoints. The future of electronic aids to daily living. Am. J. of Occupational Therapy, 56 (1), 107-109. Médical Intelligence, A Simple Solution to http://www.medicalintelligence.ca/en/columba.html Prevent Disappearance, 2005, Neuropage, The Oliver Zangwill Centre Princess of Wales Hospital, Cambridgeshire, UK http://www.neuropage.nhs.uk/ Patterson, D. J., Liao, L., Gajos, K., Collier, M., Livic, N., Olson, K.,Wang, S., Fox D. & Henry Kautz (2004), Opportunity Knocks: a System to Provide Cognitive Assistance with Transportation Services , Proceedings of UBICOMP 2004: The Sixth International Conference on Ubiquitous Computing , October 2004, LNCS 3205 , Springer-Verlag , Nigel Davies, Elizabeth Mynatt, Itiro Siio, eds., pp. 433-450. Pigot H., Savary, J.P., Metzger, J.L., Rochon, A., & Beaulieu, M. (2005) Advanced Technology Guidelines to Fullfill the Needs of the Cognitively Impaired Population. 3rd International Conference on Smart homes and health Telematic. Canada. (Under press) Pigot H., Lefebvre B., Meunier J.G., Kerhervé B., Mayers A., & Giroux S. (2003), The role of intelligent habitats in upholding elders in residence, 5th international conference on Simulations in Biomedicine, Slovenia, April 2-4, 2003. 497-506. Pollack, M. E. (2004). Special Committee on Aging United States Senate Hearing on : Assistive Technology for Aging Populations. http://www.eecs.umich.edu/~pollackm/Pollackweb_files/senate-testimony.pdf. Shriqui, C.L., & Nasrallah, H. A. (1995). Contemporary Issues in the Treatment of Schizophrenia. Gilmore Academic Psychiatry. Szymkowiak, A., Morrison, K., Prveen Shah, P.G., Evans, J.J., & Wilson B.A. (2005) A memory Aid with Remote Communication using Distributed Technology. Personal and Ubiquitous Computing. 9: 1-5. Visions (2005) The visions system. http://www.thevisionssystem.com/, 2005. Weiser, M., (1991), The Computer for the 21st Century, Scientific American.