esanté et quelques projets Henning Müller
Informatics in health Biomedical informatics Medical informatics Bioinformatics ehealth e-health Health informatics 2
A larger definition (wikipedia) Electronic health records: enabling the communication of patient data between healthcare professionals (interoperability, health cards, ) Telemedicine: physical and psychological treatments at a distance Consumer health informatics: use of electronic resources on medical topics by healthy individuals or patients Health knowledge management: medical journals, best practice guidelines or epidemiological tracking (LinkedLifeData) Virtual healthcare teams: consisting of healthcare professionals who collaborate and share information on patients through digital equipment mhealth or m-health: includes the use of mobile devices Medical research using Grids: powerful computing and data management capabilities to handle large amounts of heterogeneous data Healthcare Information Systems: often refer to software for appointment scheduling, patient data management, work schedule management and other administrative tasks surrounding health 3
A revolution in healthcare? 4
et en Suisse http://www.e-health-suisse.ch/ Web page is always updated with last initiatives Much work has been going on for several years June 2007: ehealth strategy for Switzerland was decided Core groups on specific topics Larger groups involving all interested stakeholders There are now: Technical and semantic standards An organizational vision and definition of processes Ideas for a legal framework Ideas of certification and accreditation 5
A legal framework An ehealth law is under preparation Future law on the electronic patient record (EPDG-LDEP), in the parliament in 2013 Collaboration between the confederation and the cantons (a long/slow process) Companies need to follow the standards Otherwise there will be no access to the markets Medical software is a medical device Security needs to be assured, certification 6
Pilot tests in the cantons e-toile in Geneva More information at http://www.e-toile-ge.ch Data accessible in decentralized form, confidentiality and security are regarded as extremely important Patient can decide on access with a physician Data access only with patient and professional card Rete sanitaria in Tessin Health network started in 1999 Partial evaluation by ehealthsuisse 7
esanté à Sierre Many ehealth activities since 2007 ehealth unit since 2010 20 persons and three professors M. Schumacher, A. Cotting, H. Müller Several types of projects EU FP7 projects (Khresmoi, PROMISE, WIDTH, VISCERAL, MD-Paedigree, Commodity12, ) FNS projects (MANY, NinaPro, ) CTI, Hasler, COST, HES-SO, NanoTerra, mandates 8
ehealth topics in Sierre Health data management and interoperability Semantic interoperability for document exchanges Health information retrieval Medical image analysis and retrieval Analysis of big data Clinical decision support On temporal data using intelligent agents Combining multimodal data of various sources Using signal and image data with other sources 9
Motivation for image management An image is worth a thousand words Medical imaging is estimated to occupy 30% of world storage capacity in 2010! Mammography data in the US in 2009 amounts to 2.5 Petabytes Riding the wave how Europe can gain from the rising tide of scientific data, report of the European Commission, 10/2010. 10
and big data? 1GB of data 1px = 10 MB typically used 11
Monthly production of hospitals (4 TB) Georg Langs www.cir.meduniwien.ac.at 12
Yearly production (~48 TB) 13
Big data challenges and opportunities Signal data in the images needs to be mapped to semantic information Reduce amount of data to keep accessible Get information for decision support Information of interest can be extremely small Simple and efficient tools are required And these might work better on big data Many rare diseases could be analyzed These are difficult as people do not know them, they are missed and incorrectly treated 14
VISCERAL EU funded project (2012-2015) HES-SO, ETHZ, UHD, MUW, TUW, Gencat Organize competitions on medical image analysis on big data (10-40 TB) All computation done in the cloud, collaboration with MS Identifying landmarks in the body Finding similar cases Annotation by experts 15
Khresmoi 4 year, 10 000 000 budget 16
Khresmoi goals Trustable information adapted to user groups All tools as open source Extract semantic information from all sources LinkedLifeData 17
Current status Project at the beginning of year 3 of 4 years Half-time User tests have started among the three user groups (much feedback on prototypes expected) Different types of interfaces Eye tracking Implement changes to adapt to the user groups 18
NinaPro Funded by the FNS, lead by IDIAP Acquisition of signal data from healthy and hand-amputated persons Improve control of prostheses through the signal data, to catch up with robotics Learn better and fast! Share the data to advance research 19
Medicoordination 20
DiagnosticAid CTI project Goals: Give to general practitioners a system for decision support based on clinical practical guidelines Obtain all the necessary data from the other care partners and include them into the medical record MediWay 21
Conclusions ehealth concerns all of us Health is becoming increasingly e Much practical work and many research projects exist in Switzerland Ranging from robotics, via images and signals to analysis of all types of data Many opportunities exist for public-private partnerships From start ups towards collaborations with larger companies 22
Questions? More information can be found at http://aislab.hevs.ch/ http://medgift.hevs.ch/ http://publications.hevs.ch/ http://www.khresmoi.eu/ Contact: Henning.mueller@hevs.ch 23