People with Parkinson’s disease (PD) suffer from motor and cognitive impairments that severely impact mobility, fall risk and multiple key aspects of functional independence. Until recently, treatment goals focused almost exclusively on symptom relief but exciting recent work by the scientific community, including also CuPiD partners has demonstrated that motor learning and rehabilitation principles can be effective even in the case of presence.

It is critical to make these rehab-like therapies accessible to patients in their home-setting since they need continuous training, as PD is a chronic neurodegenerative disease. In addition, optimal rehabilitation of a neurodegenerative disease like PD requires personalised training paradigms that patients can integrate into their everyday routine in their own homes and use for many years. Ongoing, long-term treatment in a clinical setting is not feasible, cost effective or something that patients are likely to comply with year after year. CuPiD is designed to meet this challenge. The project partners are developing an ICT-enabled solution for motor learning in patients with PD in their home setting, tailoring the solution to target mobility, cognitive function and debilitating PD symptoms such as freezing of gait and gait impairments.

Key components of the CuPiD solution are:

• a home-based rehabilitation system (based on unobtrusive wearable sensors, on-board intelligence for real-time feedback on the user to correct the movements, exer-gaming and modular, multi-modal restitution interfaces);
• an intelligent telemedicine infrastructure (tele-rehabilitation service) for remote monitoring and supervision of the rehabilitation program by a clinician.

The integrated, easy-to-use system will have a huge, beneficial impact on the therapeutic treatment of PD, empowering patients to improve their health-related quality of life in the comforts of their own home. At the same time, the project envisions that the ultimate costs of treatment will be reduced and the healthcare burden dramatically lowered when the unique CuPiD system becomes available.

One of the peculiarities of the project is the inclusion of almost 1 year of trials of the ICT-based solutions on patients, in their home environment, in the two clinical centres involved in the study. The trials are designed to include 6 weeks of continuous patient-specific training and a follow-up evaluation and the study aims at both validating the prototypes in terms of functionality, usability, effectiveness and reliability, but also at showing evidence of rehabilitative efficacy, toward an extended clinical validation, foreseen in a possible development of the CuPiD project.

Policy Context

CuPiD is powered by an eight member consortium led by the University of Bologna and is an FP7 ICT Collaborative Project.

Description of target users and groups

Patients suffering from Parkinson’s disease and healthcare professionals taking care of them.

Description of the way to implement the initiative

CuPiD is a three year EU project whose aim is to provide technology-driven personalised rehabilitation exercises for people with Parkinson’s disease at home.

CuPiD’s objectives are the following:

• Produce clinical guidelines for developing tailored rehabilitation programs using technology;
• Create a home-based rehabilitation system (wearable sensors and local processing);
• Build a telemedicine infrastructure for remote supervision of the rehabilitation.

The CuPiD project consortium will design, develop and test in the field an overall service, based on personal health system technology, for at home rehabilitation and training of major motor impairments caused by Parkinson’s disease.

The guiding principles which will inspire CuPiD rehabilitation are:

• sensory augmentation and substitution (i.e. the use of devices that assist a functional human sense and the use of one human sense to receive information normally received by another sense, respectively);
• multi-sensory stimulation via multimodal restitution interfaces;
• integration of motor and cognitive strategies to support motor learning or re-learning;
• delivering services in the patient’s home environment.

The CuPiD project will pilot three complementary rehabilitation services:

• Real-time audio-feedback for gait rehabilitation 
• Rehabilitation for freezing of gait 
• Exer-gaming training

The rehabilitation service which will be developed in CuPiD is intended for a delivery outside hospitals and care centres. It will be designed to be ubiquitous, so that it can be easily delivered at the patient homes, or in some specific cases (e.g. gait training, which needs adequate space for performing long-distance walking) outdoors or in their preferred environment.

An example of an audio feedback exercise for gait would be walking. A person with Parkinson’s disease would place inertial sensors on their shoes and then wear earphones attached to a smartphone. They walk for a short period to calibrate the system to their gait. They are then able to walk freely. If their gait falls outside of parameters configured for their exercise they receive audio feedback to prompt improvements.

For freezing of gait, the person with Parkinson’s disease would wear the same sensors on their legs. As with the previous example the user wears earphones attached to a smartphone. CuPiD software on the phone analyses accelerometry from the sensors and detects, or preferably predicts, an episode of “freezing” in which gait stops. Audio cueing is then provided which helps the user overcome freezing and walk normally.

A sample of Exer-Gaming is given by the task of reaching from a seated position to “touch” virtual balloons. The user views a human figure, an avatar, on the CuPiD Home system screen. The user wears CuPiD sensors on their arms and this enables the avatar to mirror their movements.

By performing the correct movements the user can cause the avatar to reach, touch and burst balloons in the game.

The three different rehabilitation services are developed following criteria based on the literature, clinical experience and direct involvement of patients. In addition, in the second stage of the project, a large group of patients is involved for experimental sessions, using the CuPiD service for training in their home environment for 6 weeks, with tailored solutions according to clinicians’ prescription. Outcomes of the sessions will be used for continuous optimisation and design of the prototypes, toward their engineering and toward a possible extended clinical validation, foreseen as possible exploitation, beyond the framework of the project.

Technology solution

Technologically CuPiD provides four complementary systems: three rehabilitation services for Parkinson’s disease and a telemedicine infrastructure which allows the delivery of each service to the patient’s home. The three rehabilitation services are:

• Audio feedback training
• Assistance with Freezing of Gait
• Exer-Gaming

Technology choices were made on the basis of:

1. Functionality: selecting technologies which supported functional user requirements. For example, the home system for people with Parkinson’s must support a touch screen, the mobile systems must run on a smartphone.
2. Existing technical investments: this was only significant in the case of Exer-Gaming in which IBIT have a mature existing system for delivering “serious games” with defined interfaces
3. Commercial viability: since CuPiD supports the large scale delivery of services to patient homes, the technologies used for the home systems should be low cost.
4. Commercial acceptability: since CuPiD supports clinicians to supervise patients remotely working in a hospital setting, the technology chosen for the hospital system must be familiar to hospital IT departments.
5. Flexibility: use of standards and open source software keeps open the possibilities of re-use of the technology in different contexts and minimises the perception that users are being locked into a system.

Our approach has been to optimise technology selection according to these five principles for each service independently and then integrate into a unified system.

For Exer-Gaming, the exercises require an Internet connection to a games server and a thick client at the patient home. The client runs on Linux (Kibuntu), which also allows deployment without licence fees. For CuPiD we are using standard web services to achieve single sign-on and information sharing between the games server and the CuPiD clinical database.

For audio biofeedback the requirements are portability and ease of use. For example, a user should be able to put a sensor on their shoe, a smartphone in their pocket and get support for walking. Technology choices here are restricted. CuPiD has developed its own sensors (see below) and connected these to Android smartphones using Bluetooth standard protocols. Android was chosen since it meets the functional requirements, is low cost and widely acceptable.

Providing support for freezing of gait is technologically similar to providing biofeedback. In both cases a simple, light system which is low-cost and easy to use is required. Therefore a combination of Android phones and CuPiD sensors was used here too. This is to be encouraged rather than proliferating technologies unnecessarily.

As part of the project, custom inertial measurement sensors have been developed by Exel using microchips provided by ST. The resulting sensors give performance and flexibility at a lower cost than their competitors. The sensors use Bluetooth standards and are described on the website: http://www.cupid-project.eu/sensor_product_specification

Technology choice: Proprietary technology

Main results, benefits and impacts

Technology

The three CuPiD services are delivered remotely via a telemedicine infrastructure. All components in this infrastructure are agnostic to the services they are delivering and therefore it can be extended beyond CuPiD. The components are a CuPiD Home System, a data exchange mechanism and a CuPiD Clinical System.

The CuPiD Home System provides a touch screen UI (whose language can be changed through configuration), authentication, user activity logging, error reporting and remote support. In order to keep the unit price of the CuPiD Home Systems low and for compatibility with the Exer-Gaming client, a Kibuntu operating system has been chosen. This launches Firefox in Kiosk mode on an All-In-One PC to provide one “box” which is operated via a touch screen. The user can select and launch any service listed in the configuration.

Transparent and reliable data exchange is achieved through a store & forward network. The system handles data sets as files (with interpretation of file content being left to the individual services). A live connection between clinician and patient is not supported.

The CuPiD Clinical System is a secure database and browser-based application. For this Microsoft technologies were used to optimise familiarity to the end users.

For the integration, standard technologies have been used throughout including Bluetooth, SSL, and WebDav. The telemedicine infrastructure will be released as open source to allow any project needing to deliver home-based patient services from a clinic (without live-patient connection) to benefit.

Outcomes

The overall outcome will be the establishment of feasibility and efficacy of a closed-loop based rehabilitation protocol to be performed at home by subjects with Parkinson’s disease. CuPiD will develop innovative rehabilitation based on new technology and the patient’s needs.

Expected outcomes will include the following:

• Demonstrate and quantify the influence of training on PD. There is still a huge matter of research and discussion in the scientific community, with clear evidences but already with enormous gaps to cover;
• Provide evidence of efficacy of therapy by means of motor-learning processes;
• Provide an ICT-based tool to administer therapy in the home environment;
• Allow research on the basic mechanisms of training in people with PD and the correlation between context (i.e. medication intake, stress, location, attention level) and motor performance through the large amount of data gathered will;
• Development and validation of new algorithms for feedback;
• Provide an innovative Personal Health System with on-board processing enabling multi-modal feedback and multi-faceted training;
• Guarantee usability and personalisation of rehabilitation, empowering the end-user to self manage their training at home.
• Contribute to the use of standardisation in eHealth.

Return on investment

Return on investment: Not applicable / Not available

Track record of sharing

A major outcome from the CuPiD project will be clinical exercise protocols, quantified results and qualitative evidence concerning the efficacy of closed-loop exercise for people with Parkinson’s disease.

CuPiD partners have presented papers or posters at conferences (e.g. SIAMOC 2012, 2013, UBICOMP2013, MEDICON 2013, ESMAC 2013, Body Sensor Networks, King’s Fund International Congress on Telemedicine, ICAMPAM, International Conference on Neurorehabilitation 2012, AAL, Pervasive Health2012). CuPiD has made individual presentations (e.g. to the Telecare Services Association, the Frankfurt Parkinson’s support group, the Department of Primary Health, Oxford University and at University Hospital, Ontario) and opened data (Freezing of Gait accelerometry data published on physionet).

CuPiD is open-sourcing the telemedicine application and sharing has already begun with the Movement Sciences Group at Oxford Brookes University. We will use a public code sharing site such as GitHub to publish the code.

Lessons learnt

The three main lessons are:

1. Research needs to be patient-centred:
   Exercise protocols and equipment that do not take into account the special needs of users such as people with Parkinson’s disease will not work. CuPiD has benefited from workshops with users run in Leuven and Tel Aviv and used these results to guide both the development of exercise protocols and to make decisions about technology.
2. Clinically, research into the long-term potential for targeted closed-loop rehabilitation protocols is needed:
   Technology may be a key element to this regard. This is an area of great potential but still needs several research and standardisation. Because Parkinson's disease is a progressive neurological disorder with increasing disability over time, the merits of rehabilitation have been questioned for many years.
   However, in the past two decades there has been a shift in this concept due to evidence from research studies demonstrating the ability of motor learning in PD, as well as improvements in function as a result of training. In fact it has been shown that pathways in the brain involved in Parkinson’s disease (the basal ganglia) may be capable of plasticity, and their activity patterns may be partly corrected with appropriate intensive training. (Fisher BE et al. “The effect of exercise training in improving motor performance and corticomotor excitability in people with early Parkinson's disease”. Arch Phys Med 2008; Tajiri N et al. “Exercise exerts neuroprotective effects on Parkinson's disease model of rats”. Brain Res. 2010).
   Because CuPiD is a possible provider of solutions for customised rehabilitation protocol without excessive costs, it might become a tool to really improve both the quality of life of patients and advancements in scientific and clinical research.
3. For telemedicine the routes to market are immature and standards for contracting and delivering services need development:
   CuPiD is testing placing telemedicine services in an online system used for contracting home-based care services. This system is already in use in four UK local government areas. This exposes the need for credibility and acceptability amongst clinicians and patients as emphasised in the EC Working Paper on telemedicine (2009) - European Commission Staff Working paper “Telemedicine for the benefit of patients, healthcare systems and society”: http://ec.europa.eu/information_society/newsroom/cf/document.cfm?action=display&doc_id=622

Scope: International