ARMOR* (ARMOR)

Published on: 11/10/2013
Document

* Full title: Advanced multi-paRametric Monitoring and analysis for diagnosis and Optimal management of epilepsy and Related brain disorders


Epilepsy, the propensity for recurrent, unprovoked epileptic seizures, is the most common serious neurological disorder, affecting over 50 million people worldwide. Epileptic seizures manifest with a wide variety of motor, cognitive, affective, and autonomic symptoms and signs and associated changes in the electrical activities of the brain (EEG), heart (ECG), muscles (EMG), skin (GSR), as well as changes in other important measurable biological parameters, such as respiration and blood pressure. Their recognition and full understanding is the basis for their optimal management and treatment, but presently is unsatisfactory in many respects. Epileptic seizures occur unpredictably and typically outside hospital and are often misdiagnosed as other episodic disturbances such as syncope, psychogenic and sleep disorders, with which they may co-exist, blurring the clinical presentation; on the other hand, costs of hospital evaluation are substantial, frequently without the desirable results, due to suboptimal monitoring capabilities.

Reliable diagnosis requires state of the art monitoring and communication technologies providing real-time, accurate and continuous brain and body multi-parametric data measurements, suited to the patient's medical condition and normal environment and facing issues of patient and data security, integrity and privacy.

This project will manage and analyse a large number of already acquired and new multimodal and advanced technology data from brain and body activities of epileptic patients and controls (MEG, multi-channel EEG, video, ECG, GSR, EMG, etc.) aiming to design ARMOR, a more holistic, personalised, medically efficient and economical monitoring system.

New methods and tools will be developed for multimodal data pre-processing and fusion of information from various sources. Novel approaches for large scale analysis (both real-time and offline) of multi-parametric streaming and archived data will be introduced to discover patterns and associations between external indicators and mental states, detect correlations among parallel observations and identify vital signs changing significantly. Moreover methods for automatically summarising results and efficiently managing medical data will be developed. ARMOR will incorporate models derived from data analysis based on already existing communication platform solutions emphasising on security and ethical issues and performing required adaptations to meet specifications. Special effort will be devoted in areas such as data anonymisation and provision of required service.

ARMOR will provide flexible monitoring optimised for each patient and will be tested in several case studies and evaluated as a wide use ambulatory monitoring tool for seizures efficient diagnosis and management including possibilities for detecting premonitory signs and feedback to the patient.

Policy Context

The ARMOR Project is partially funded under the 7th Framework Programme by the European Commission. It is developed by 8 organisations of 5 European Countries (Greece, Spain, Cyprus, Germany and United Kingdom).

Description of target users and groups

The ARMOR project addresses the needs of the epileptic patient and healthcare professional, aiming at the design and development of a framework for the monitoring and analysis of epilepsy-relevant multi-parametric data, in the context of Personal Health Systems (PHS).

Description of the way to implement the initiative

ARMOR provides on-line and off-line analysis of data with the help of medical databases and patient's medical file for the purpose of assisting diagnosis, prognosis and treatment as well as for predicting and classifying seizures. New methods and tools are developed for multimodal data pre-processing and fusion, real-time and offline data mining of multi-parametric streaming and archived data to discover patterns and associations between external indicators and mental states, lag correlations, motifs or outliers (vital signs changing significantly), automatic summarisation of results and efficient medical context data management.

Offline analysis mostly concerns the accurate diagnosis of seizures (recognition and discrimination between epilepsy and non-epileptic paroxysmal events) and the identification of various risks of disease development, recurrence. Generally it provides assistance to health professionals in suggesting appropriate treatment and evaluating its effectiveness. Computer-assisted diagnosis services assist the decision-making process when a medical condition needs to be treated, providing plausible explanations related to the abnormal combination of vital sign values. These explanations are mostly based on the outcome (conclusions and guidelines) of the offline data analysis module and are enhanced with the personalised patient health profile together with models created for each different type of epilepsy. Information obtained by the analysis of patient's health condition and their environment is used in the decision support system.

Real-time (online) analysis is performed on multi-parametric stream data to detect signals beyond the limits, identify seizure premonitory signs, discover typical patterns of activity followed by seizures, detect atypical patterns of activity/behaviour based on models that are created. Methods for combining the information extracted from the patient's health profile with real-time sensor data are developed, aiming to provide accurate information to the online decision support system and to deliver appropriate alerts to the health professional.

Image removed.

State-of-the-art measuring system and flexible coordination/communication platform are used, allowing long term monitoring, medical management and decision support suited to the patient's medical condition without disrupting the patient's life pattern and normal environment while facing issues of patient and data security, integrity and privacy.

 

ARMOR's main objectives can be viewed from three main perspectives:

  • Clinical:
    • Facilitate and increase the yield of diagnostic, treatment and follow-up practices in epilepsy;
    • Support health care professionals in their decision making and provide accurate signaling of risks;
    • Reduce epilepsy related management costs;
    • Advance the technology of Personal Health Systems (PHS) suitable for chronic diseases of multifactorial causes and unpredictable expression of their symptoms.
  • Medical / Theoretical:
    • Increase understanding of the epileptic seizure and eventually of epilepsy;
    • Increase understanding of the other non-epileptic paroxysmal events (NEPE) and their underlying mechanisms that is currently poor;
    • Understand possible relationships between epilepsies and the various types of NEPE.
  • Information and Communication Technologies (ICT):
    • Design and develop adequate measuring/monitoring methodologies and systems facilitating the respective medical objectives;
    • Develop multi-parametric data processing, management and analysis tools, both real-time and off-line, to better understand epilepsy;
    • Adopt and adapt prominent coordination/communication platforms providing robust and flexible end-to-end communication and assure security and privacy on sensitive medical data.

Technology solution

Technology choice: Proprietary technology

Main results, benefits and impacts

  1. ARMOR could offer an EEG diagnostic and LTM service achieving in-hospital quality standards, and addressing conventional “routine” clinic-based EEG service purposes, at reduced cost and increased geographical availability, likely to be of interest to commercial and public-sector healthcare providers.
  2. ARMOR could offer an ambulatory EEG service achieving in-hospital quality standards, and addressing conventional “routine” clinic-based EEG service purposes, with increased geographical availability, and with enhanced capability through other parameters of data collection, especially video. Although it is uncertain whether there would be cost savings relative to conventional ambulatory EEG, it is likely to be of interest to commercial and public-sector healthcare providers.
  3. ARMOR would be able to build on the success of Kings@home, to provide a much more extensive range of services, as described in the ARMOR scenarios, and to integrate with miniaturised multiparametric monitoring devices, and online tools and services. This may be of interest to healthcare providers, and is likely to be of interest for direct marketing to consumers (e.g. for monitoring emergency situations). DigiTrace provides a model of successful direct marketing of a diagnostic service to consumers as well as clinicians. The success of motion detection devices, despite very poor evidence for their effectiveness, provides some encouragement that there is a consumer market ready for an effective seizure-detection service.
  4. ARMOR could seek to integrate, in the future and if there is apparent competitive advantage, successful outcomes of the video seizure detection project at SEIN, successful outcomes of the EPILEPSIAE seizure-prediction project, and successful outcomes of the PERFORM Parkinson’s Disease remote monitoring project.

Track record of sharing

The ARMOR project has exploited different channels for disseminating the project results, achievements and lessons learnt, including the website http://armor.tesyd.teimes.gr/, a twitter account, leaflets, poster and newsletters distributed at several events, with scientific community participation including both ICT and health experts.

  • An important example is the attendance of one of the partners (AAISCS) to the 1st European MEG Society Teaching Course in Chiemsee, near Munich in Germany.
  • Another important training event where ARMOR was presented is the ERASMUS IRIS International Summer School: Cellular Technologies & Services. IRIS took place at the premises of Technological Educational Institute of Western Greece, Computer and Informatics Engineering Department at Antirion, Greece, from 8-19 July to July 2013.
  • Collaboration with sister projects, networking and information sharing was also the aim of the participation of one ARMOR partner (AAISCS) to the Michelangelo Concertation meeting – 25 September 2013, Brussels.
  • Finally the ARMOR project has been selected as an exhibitor in the ICT2013 in Vilnius (6-8 November 2013). At that event it will be possible to demonstrate the functionalities and features of the whole ARMOR system. The ARMOR members in the stand will put to themselves different variety of sensors (EEG, Activity, ECG, etc.). The sensors will acquire the information from the sensors, pre-process, encrypt and monitor the parameters in real-time. All the data captured will be sent to a database where both patients (and their relatives) and doctors will be able to consult in real-time the values captured by sensors, as well as in anomalous situations. The demo will be performed also on the visitors who agree to put on the sensors. More than 4000 researchers, innovators, entrepreneurs, industry representatives, young people and politicians are expected in Vilnius.

Lessons learnt

  1. Enhance communication security and performance through encryption and compression algorithms: The encryption algorithm is provided offering highly competitive performance metrics, including speed, power consumption, and silicon area demands. Innovation in the compression algorithms area included the proposal of adequate algorithms and the efficient hardware implementation of them so as to require minimum resources.
  1. Importance of reduced cost and increased geographical availability: The home gateway maintains permanently a link to a central server (PHR) where all data from modalities are stored in automatic way, by schedule or triggered by an event signal coming from the online algorithms or sensors. In this way, sensor data is always almost ‘on-line’ for doctors and/or caregivers. Home gateway provides a means of communication of important events from the patient to caregivers and/or doctors, offering the way to maintain a direct link between them. Based on the events reported from the home gateway, and the protocol defined, doctors can provide instant attention to those events/patients that requires special type of assistance.
  1. Being user-friendly: Minimising/reducing the required number of sensors to extract critical results is the first step to achieve a better user acceptance. The ARMOR project proposes a new way of looking at the Personal Health Record (PHR) as an enabling technology for supporting medical research on data coming from the large patient base. Approaches to analysing individual health condition based on rules derived from data of a group of patients is also an innovative approach.
Scope: International