REMPARK: Personal Health Device for the Remote and Autonomous Management of Parkinson’s Disease (REMPARK)

Published on: 09/09/2013
Document

The specific and ultimate goal of the REMPARK project is to develop a Personal Health System (PHS) with closed loop detection, response and treatment capabilities for the management of Parkinson's Disease (PD) patients at two levels:

  • At the first level, the project will develop a wearable monitoring system able to identify in real time the motor status of the PD patients, and evaluating ON/OFF/Dyskinesia status, with sensitivity greater than 80 % and specificity greater than 80 % in operation during ambulatory conditions and will also develop a gait guidance system able to help the patient in real time during their daily activities.
  • At a second level, the intelligent analysis of data provided by the first level, supported with a disease management system will allow the neurologist in charge to access accurate and reliable information to decide about the treatment that best suits the patient, improving the management of their disease, in particular to adjust so called therapeutic window.

More about the objectives and goals of the project can be found on the following video:

Policy Context

The main problems associated with Parkinson disease are the motor fluctuations suffered by the patients. The motor fluctuation observed (tremor, bradykinesia, postural alteration, alteration of gait speed and stride length, freezing of gait and falls) depend on the patient, the stage of the disease and its evolution.

Medication is used to mitigate the set of symptoms, and the effectiveness of this will depend on the patient and the relative evolution of the disease in this patient.

Neurologists try to set the medication dose in order to avoid the presence of motor fluctuations as much as possible. The problem that neurologists currently face is the lack of quantitative information on the intensity of the symptoms and their duration. Neurologists would be able to manage the disease more effectively if this information could be provided.

Description of target users and groups

The targeted users or group of users are those related and obtaining benefits from an improved management of the disease. Patients are the primary users with the main advantage of quantitative measurement and record of motor symptoms evolution, helping in this way the improvement of the medical management and supervision of their disease. Second main group is the neurologists group with an improved tool to follow the evolution and management of the PD patients.

Description of the way to implement the initiative

To achieve the described objectives, a telemedicine enabled Personal Health System (PHS) will be developed for the remote management of Parkinson's Disease in its medium and advanced stages. Two levels compose the PHS:

  • The first level corresponds to the Body Area Network (BAN) and acts in the short-term. It is composed by the sensors, the actuators and the smartphone, which also acts as GPS for providing context-aware information and as an interface for the PD patient, to record their direct feedback especially regarding the non-motor symptoms. Communications among the elements of the BAN will be via low energy Bluetooth (4.0), allowing enough autonomy for the patient. This first level will work autonomously, constituting a closed and automatic loop. This level of the system will be auto-adaptive by means of a constant evaluation of the actuator's effect, correcting its behaviour in the short term. Different configurations are possible for different patients. For instance, some patients may react better to auditory cueing while others may react better to haptic cueing (provided by the FES device). The use of FES as step initiating device will be an option for those patients suffering frequent episodes of FOG, while others may not need it because they do not experience FOG episodes, or because the chosen cueing system is enough to prevent the FOG episodes. The adjustable drug delivery system will be investigated and tested in subsequent medical projects and the objective is to include it for those patients suffering from unpredictable OFF periods.
  • The second level acts in the medium-long term and constitutes a closed semi-automatic loop, as it allows the intervention of medical professionals. The system will be able to send data to the server of the relevant health service provider, allowing the patient's neurologist to regularly follow the evolution of the patient's disease in a more effective manner, as well as being able to make better informed decisions about the adjustment of the pharmacological treatment of the patient, a key issue in management of this disease. Data on the server will be automatically included in an Electronic Health Record (EHR). The correct intelligent data treatment will help to evaluate and predict the evolution of the disease of a particular patient.

On this specific architecture, two different groups of algorithms will be running. The first group will be located in the level 1 (BAN), and it will be responsible for the ON/OFF detection and the implementation of the gait guidance system. A second group of algorithms will be the responsible for the implementation of the Rule engine at the server level.

The REMPARK objectives will be achieved using the following approach:

  • Real-time motor status identification under ambulatory conditions will be achieved using two wearable inertial, electronic, intelligent sensors (one placed in the waist and one in the wrist).
  • The intelligence of the artificial system will be achieved using algorithmic encapsulation of the acquired data in a large database constructed in the first phase of the project. The intelligent system will be required to correctly identify ON, OFF and Dyskinesia states, to detect Freezing of Gait (FOG) episodes and to identify and register falls.
  • The gait guidance system and FOG treatment will be activated as a consequence of the motor status identified, and it will be mainly based on a self-adaptive auditory cueing system. A self-adaptive haptic cueing system, based on electronic actuation will be also developed and evaluated.
  • A Smartphone User Interface will facilitate communication between the patient and the clinical team on issues such as medication, quality of sleep etc. It will also permit the REMPARK system to send specific prompts to the patient asking for the validation of events detected by the REMPARK system. The interface will also be used to facilitate medical tailored tests to assess the patient’s clinical evolution.
  • The REMPARK server will permit the clinical team to send appropriate acquired and processed data to the repository. Eventually, these data will be integrated with the patient’s Electronic Health Record (EHR). The server will contain a disease management tool based in a rule engine developed within the framework of the project.

Image removed.

Technology solution

Technological solution, as it has been briefly announced, is based on specific systems built around standard subsystems. Main parts of the system are:

  • UPC owns the specific movement sensor with embedded algorithms running inside in real time and it is an improvement of a specific earlier design, mainly based on accelerometers and gyroscopes.
  • UPC based on the extracted knowledge from the constructed database develops the embedded algorithms. The database has been a very important work along the second period of the project with the cooperation of the medical partners involved in the consortium.
  • The interface with the user is done using a Smartphone with specific software application developed by FhP in Portugal. This interface is used for many purposes: communication with the user, automatic feedback from the server, automatic generation of an auditory queuing, generation and submission of tests for non-motor symptoms validation.
  • The server with the data and the medical rule engine will be a standard machine with the specific algorithms and rules installed.
Technology choice: Proprietary technology, Standards-based technology

Main results, benefits and impacts

The REMPARK system will continuously provide data to the central server concerning the PD status and associated parameters of its patients. These data include not only sensor measurements, but also objective metadata and subjective symptomatic information provided by the patients using the Smart Phone interface.

It is thus expected that REMPARK will in this way enhance the active and regular participation of the patients in care processes. Doctors will be able to provide feedback to the system, adjusting the PD assessment and treatment regime individually for each of their patients monitored by REMPARK.

Some strategic results expected

• A large labelled database containing ambulatory activity of PD patients will be available.

• Knowledge extracted from the labelled database will be embedded in the implemented intelligent models.

• A self-adaptive cueing system will be developed for FOG overcoming.

• High-level intelligent data processing on the REMPARK server will be implemented for assisting doctors with the management of the disease in the patients.

Strategic impacts

• Expected reduction of the hospitalisation of PD patients using the system.

• Expected improvement in the management of the disease.

• Precise detection of ON/OFF/Dyskinesia states on a large sample of patients.

• Improved medical knowledge on Parkinson’s disease through the quantitative evaluation of associated motor problem states.

Economic benefits

• While it may be difficult to evaluate the impact of REMPARK on the indirect costs of Parkinson’s disease, the achievement of better control of the ON/OFF periods with PD patients, should result in a significant percentage of patients requiring less assistance and should consequently increase.

Track record of sharing

This has not been established yet.

Lessons learnt

REMPARK project is still in an intermediate progressing state. Work is advancing according to the main schedule, with minor changes.

Main lessons learnt will be fully available at the end of the project, after the complete experience with the PD patients, the medical involved teams in the project and all the participating partners (mainly with technological skills). Current lessons learnt include:

  1. When organising a research project involving patients and users at different levels, it is very important to schedule all the activity. In REMPARK case it has been crucial the definition of the method for data acquisition and it has been only possible with the correct cooperation of all involved actors.
  2. All the issues related with regulation and Ethical Committees are compulsory and very time consuming. In REMPARK it has been an important nucleus of activity during the first period of the project.
  3. When an activity is involving technological competences and medical protocols, it is essential to organise mutual training sessions in order to improve cooperation and working in a successful organisation.
Scope: Pan-European