1. Real-time telemedicine
in paediatric cardiology
Riccardo Triunfo, Francesca Frexia, Francesco Cabras, Carlo Buttu, Valeria Lecca, Stefano
Gessa, Sabrina Montis and Roberto Tumbarello
eTELEMED 2013
The Fifth International Conference on eHealth, Telemedicine, and Social Medicine
February 24 - March 1, 2013 - Nice, France
3. CRS4
CRS4: CENTER FOR RESEARCH, DEVELOPMENT AND
ADVANCED STUDIES IN SARDINIA
‣ Interdisciplinary research center focused on computational sciences
‣ Located in the POLARIS Science and Technology Park (Pula, Sardinia, Italy)
‣ Operational since 1992
‣ Scientific/industrial research, technological development and higher education
‣ RTD staff of ~170 people
4. CRS4
STRENGHTS AND FOCUS
‣ Key strengths:
‣ Infrastructure
‣ Research and development in enabling
technologies
‣ Direct experience in the application context
‣ Primary focus is on
‣ Energy & Environmental sciences
‣ Information society
‣ Biomedical sciences
‣ International and national collaboration
‣ PON, EU FP7, Wellcome Trust, NIH, APL, …
‣ Technology transfer toward industry and regional
structures
‣ ENI, INPECO, IBM, NICE, GEXCEL,…
6. HEALTHCARE FLOWS GROUP
‣ Part of Data Fusion Sector
‣ More than ten year experience in research and
development in ICT application for clinical practice,
medicine and biology
‣ State of the art solutions to real medical needs
‣ Collaborations/direct involvments:
‣ Hospitals
‣ Enterprises
‣ IHE
‣ HL7 international
‣ OpenEHR consortium
‣ OME consortium
HEALTHCARE FLOWS GROUP
7. Semantic and computable management of biomedical data
Clinical domain interconnection and traceability
Telemedicine and distributed medicine
HEALTHCARE FLOWS GROUP
MAIN RESEARCH TOPICS
8. CLINICAL DOMAIN INTERCONNECTION AND TRACEABILITY
‣ Hospital Information System has to communicate with many systems (LIS, RIS, CIS,
…), different for structure and acquisition: connection one-to-one create a
complexity scaling with the square on the systems number
‣ Precise control on information improves quality and security for the patients
‣ We design different paradigms, for moving from information island to smarter
interconnection models, adopting international standards and guidelines like HL7
and IHE (eg: HL7 gateway, based on Mirth, working since 2009 in the biggest
Sardinian hospital)
HEALTHCARE FLOWS GROUP
9. SEMANTIC AND COMPUTABLE MANAGEMENT OF
BIOMEDICAL DATA
‣ The amount of data and information nowadays
available is increasing at an high rate, making
essential a robust data modeling
‣ Working on large volumes of data is intrinsically
different than working on small datasets, standard
computational tools will break
‣ Needs of semantic computable data models,
robust in time and suitable for different kind of
users (fomalism like openEHR, CDA,…)
‣ We study and realize data models to represent
and manage biomedical data (eg: model for
managing genetic data from a NIH large scale
population study)
HEALTHCARE FLOWS GROUP
10. TELEMEDICINE
‣ Recognized at international and national level as an important tool to obtain:
‣ Services of better quality and wellbeing for patients (home is where you re
cared for)
‣ Cost reduction, new care models
‣ Positive contribution to industrial sector
‣ A big part of telemedicine projects is not able to evolve from pilot studies to
tools used in real clinical contexts
‣ In Europe
‣ Telemedicine is a key point of European Digital Agenda for e-health, one of
the 7 flagship
‣ In Italy
‣ eGov 2012, telemedicine is among the prioritary sector with booking
centers, HER and e-prescription
HEALTHCARE FLOWS GROUP
11. TELEMEDICINE
‣ E-Health situation in Sardinia Region
‣ 1.6 Milion people, 11 Public Health structures
‣ Hub&spoke model promoted
‣ Considerable investments (network, infrastructures) for the creation of a
regional HIS
Project name Investment Areas
SISaR 25M€ ADT, Scheduler, Emergency, Order Entry,
Blood Banks, Pharmacy, Operating Theatres
MEDIR 10M€ GP and paediatrician network, EHT
SILUS 1.5M€ Laboratory
RTP 1.1M€ Cancer registry and teleconsultation
ANAGS 1.3M€ Federated Demographics DB
RTR NA Optical Fiber Networks
HEALTHCARE FLOWS GROUP
13. REMOTE PROJECT
RESOURCES AND ORGANIZATIONAL MODELS IN
TELE-ECHOCARDIOGRAPHY
REMOTE PROJECT
‣ The project , partially funded by Sardinian Region, proposes a model to enable
access to specialized care in remote areas using real-time telemedicine and low
cost and open technologies
‣ The clinical focus in on operator dependent diagnostic methodologies, whose
results may not be revealing when the exam is not performed by a specialist
14. ‣ Congenital heart diseases (CHDs) are the most common congenital disorders
affecting 6 to 13 ‰ live-born infants and can be detected by echocardiography
‣ The diagnosis is complex, strongly operator dependent: even a expert cardiologist
can’t perform this kind of examination, if not specifically trained
‣ Barriers rise to the access of specific exams, particularly in sparsely populated areas
because of a lack of physician with a specific skill in the pathology
‣ In Sardinia there is a high incidence (double of international frequency), but only a
tertiary center, Brotzu Hospital, with specialized expertise in the Paediatric Cardiology
Structure directed by Dr. Roberto Tumbarello
CASE STUDY: PAEDIATRIC CARDIOLOGY
REMOTE PROJECT
15. ‣ Issue: high incidence of congenital hearth diseases
‣ Resolution: virtual presence of specialists through telemedicine applications
‣ Issue: operator dependent exam
‣ Resolution: direct interactions between the operator and the specialist (real-time
voice and video chats) enables correct exam execution.
‣ Issue: lack of specialists and devices in secondary healthcare structures
‣ Resolution: the system allows better logistics, concentrating specialists in tertiary
hospitals, reducing shifts and reusing ultrasound devices heterogeneity and age
(only a video output is needed)
‣ Issue: low investments in remote districts
‣ Resolution: COTS low cost devices and open source software.
MAIN ISSUES/RESOLUTIONS
REMOTE PROJECT
16. ‣ The main objectives are:
‣ enable real-time consultations with direct interaction between clinicians involved
‣ reuse any diagnostic devices available in remote districts
‣ restrain the costs of additional material/software required
‣ lower the learning curve for the secondary care doctors
‣ facilitate teaching sessions via video conferences with many participants
OBJECTIVES
REMOTE PROJECT
19. SCHEDULING
‣ Routine
‣ scheduling system consistent with Sardinian regional scheduling system
‣ Emergency
‣ Access to consultation during availability periods
‣ Emergency calls always available
REMOTE PROJECT
20. TELECONSULTATION START
‣ The specialist plans the teleconsultation with the requiring physician (via system
scheduling or by phone for emergencies)
‣ The requiring physician starts teleconsultation in 3 simple steps (authentication,
request and start)
‣ At this point only the specialist response is needed
REMOTE PROJECT
21. WAITING FOR THE SPECIALIST
‣ The specialist starts the system in his/her laptop
‣ The specialist accepts teleconsultation in 3 simple steps (authentication, request
accepted and start)
REMOTE PROJECT
22. PROJECTION REQUEST
ECHO PARAMETER CORRECTION
PROBE POSITION CORRECTION
SCREENSHOT ACQUISITION
AUDIO FROM A CANALE DOPPLER
DURING TELECONSULTATION
‣ The physicians are connected by a audio/video channel
‣ The specialist can see in real time the echograph output and the examination scene
recorded by a camera, and can guide who is performing the diagnostic test
REMOTE PROJECT
23. DEMOGRAPHIC DATA
ANAMNESIS
PREVIOUS EXAMINATIONS(ECG, etc)
STRUCTURED REPORT
ECHOCARDIOGRAPHY
AFTER TELECONSULTATION
‣ The specialist prepares the report for the examination
‣ A structured report is used, to obtain uniformity and to automatically analyze the data
‣ The report can be exported in the most common formats (CDA included) and
integrated in Sardinian Region Ambulatory Care system
REMOTE PROJECT
31. ARCHITECTURE
TECHNOLOGY HIGHLIGHTS
‣ Main hardware and software technologies are state of the art but stable (eg. Voip,
streaming protocols,…)
‣ Central role of COTS technologies and opensource software
‣ Easy porting to different architecture
‣ Attention to system scalability
OSX Windows Linux Android IOS
LANGUAGE PYTHON PYTHON PYTHON JAVA OBJECTIVE-C
AUTH OAUTH OAUTH OAUTH OAUTH OAUTH
API QT QT QT native native
GUI QT QT QT native native
VOIP PJSIP PJSIP PJSIP PJSIP PJSIP
AV GSTREAMER GSTREAMER GSTREAMER native native
BANDA INTERNET
DATA HUB
APPLICATION SERVER
VOIP HARDWARE
DECENTRALIZZAZIONE
VOIP
32. STREAMINGS AND CONNECTIONS
‣ All the connections are encrypted via VPN channels
‣ 3 authentication levels (application, device, users) via standard streamline (OAUTH)
‣ User profiling, also according to the operational unit
‣ Secure https connection (via CA and certificates for single devices)
‣ Standard protocols used for streaming (MPEG4/H264 over RTP)
‣ SIP protocol for VOIP channel routing
‣ Audio/video transmission optimized to maintain synchronicity and low latency
‣ Controlled maximum bandwidth required for each single consultation (2.5 Mbps )
IN OUT
BASE STATION 500 kbps 2.5 Mbps
CONTROL STATION 2.5 Mbps 500 kbps
DATA HUB 500 kbps 500 kbps
TECHNOLOGY
33. EXPERIMENTATION
‣ The system was tested on 42 patients, performing the examination both in the
traditional way and via teleconsultation
‣ The experimentation involved two specialist in Paediatric Cardiology, working
directly or remotely according to an established rotation
‣ The protocol included specific steps and a pre-defined set of projection
‣ The structured report was fundamental to obtain comparable results
‣ The results for the global diagnosis were good, showing a complete match between
the two diagnosis in the 97,6% of the cases
‣ The results obtained comparing the single voices of the report are satisfying and will
be sent by the clinicians to American Hearth Association Congress
CLINICALTRIAL
34. RESULTS
‣ Valid and appreciated teleconsultation system, low cost and opensource
‣ Application for consultation management and structured reporting
‣ Connection models dedicated to the different needs of the clinical units, with
heterogeneous diagnostic devices
‣ Organizational models for the clinical unit cooperation
‣ Opportunity for patient/clinician stress reduction and efficiency
‣ Successful clinical experimentation
‣ Streaming one-to-many for educational purposes
REMOTE PROJECT
35. FUTURE DEVELOPMENTS
‣ Continuation of the research activities, to
obtain real mobility and new interaction
paradigms based on virtual reality
‣ Application in other clinical contexts with
operator dependent diagnostic methods (FAST
examination, haemodynamics,…)
‣ Connection of other centers, in Sardinia and in
the rest of Italy (interest from Bologna, Trieste
and Genoa)
‣ Software refactoring and release under
opensource licence
‣ Industrialization and diffusion
REMOTE PROJECT
36. CONCLUSIONS
‣ Technology is very important but must be an instrument
‣ User-inspired and user-driven approach
‣ Bottom-up development, but with a general point of view
‣ Attention to international standard and guidelines
‣ Attention to interoperability with existing systems
‣ Double perspective, research and real life experience
REMOTE PROJECT