40

Cloud computing applications for the healthcare industry issue 3 2014

Embed Size (px)

DESCRIPTION

 

Citation preview

1

CloudSource Editorial - Issue 3Welcome to the third issue of CloudSource Magazine, devoted to cloud computing research and development in the healthcare sector.

As healthcare worldwide undergoes a digital transformation – moving from paper reports and film-based examination products to high-definition digital content – cloud computing solutions are well positioned to alleviate the burden of the paper/film trail of medical data, streamlining operations for care-providers, hospitals, insurance companies, public-health organizations and more. More importantly, the computational capabilities afforded by cloud computing can help meet requirements for timely analyses of diverse high-definition medical records, and accurate provision of personalized care.

In this emerging healthcare environment, there are not only opportunities to be harnessed, but also obstacles to be overcome if effective and secure services are to be offered to the public.

This issue of CloudSource Magazine outlines the work of some key EU-funded initiatives, showcasing the cloud-based innovations that are enabling effective solutions for healthcare stakeholders. Among others the ClouT project, in collaboration with the City of Mitaka in Japan, explores ways to better offer cloud-based services to the elderly, by exploiting both remote-sensing data and information on city services.

ENSURE seeks to manage the exponential growth of medical data by developing the PDS object storage system; PDS enables the creation of private storage and allows for the execution of storlets to save in terms of computations that occur within the PDS elements. The PINCLOUD project deploys service-oriented architecture on the cloud, aimed at provisioning home-care telemedicine and electronic prescription, referral and learning. VIGOR++ aims at personalizing care for Crohn’s disease. A platform-as-a-service (PaaS) solution enabled by 3DNet, it has 500,000 case studies under its belt, each featuring thousands of images.

The SoftCare application, powered by the SEACLOUDS project, offers solutions for the early detection of patient symptoms, improved self-care and care-at-home as well as the creation of social networks.

Finally, DEDALUS s.r.l., the Italian software company coordinating the MIDAS project, presents a cloud-based environment for the integration of public health records (PHR) and electronic health records (EHR), advocating a new model that involves physicians, pharmacists and the public all accessessing services from an integrated system.

We wish you an informative and enjoyable read!

The CloudSource Team

2

SUCRE Coordinator, National & Kapodistrian University of Athens, Greece

Head of the Supercomputing Department at the Poznan Supercomputing Center, Poland.

President of ERCIM, U.K.

OCEAN project Coordinator, Fraunhofer Institute, Germany

EU-Japan Centre for Industrial Cooperation, Project Manager

Scientific Editor and Journalist, Australia

Coordination by Giovanna Calabrò, Zephyr s.r.l., Italy and Eleni Toli, National & Kapodistrian University of Athens, Greece.

This publication is supported by EC funding under the 7th Framework Programme for Research and Technological Development (FP7). This Magazine has been prepared within the framework of FP7 SUCRE - SUpporting Cloud Research Exploitation Project, funded by the European Commission (contract number 318204). The views expressed are those of the authors and the SUCRE consortium and are, under no circumstances, those of the European Commission and its affiliated organizations and bodies.

The project consortium wishes to thank the Editorial Board for its support in the selection of the articles, the DG CONNECT Unit E.2 – Software & Services, Cloud of the European Commission and all the authors and projects for their valuable articles and inputs.

Prof. Alex Delis,

Dr. Norbert Meyer,

Prof. Dr. Keith Jeffery,

Dr. Yuri Glikman,

Dr. Toshiyasu Ichioka,

Mrs. Cristy Burne,

Editorial Board

3

1

2

3

4

8

11

14

18

20

24

27

30

34

37

38

Editorial

Editorial Board

Table of Contents

SCAN: A Cloud-based Analytic Pipeline for Advance Cancer Prevention and

Diagnostics

ClouT: Cloud of Things for empowering the citizen clout in smart cities

A cloud-based PHR-EHR integrated environment for personalization of care

ENSURE: Long term preservation of digital data with benefits

Let’s talk about new services for patients and healthcare organizations, not cloud

PINCLOUD: INTEGRATED E-HEALTH SERVICES OVER CLOUDCELAR: Automatic,

SOFTCARE: Multi cloud-enabled platform built around the needs of elderly people

Bottom-up: Towards Supporting Personalized Medicine in the Cloud

Quantitative Medical Imaging in the Cloud: Enabling VIGOR++ with 3DNet

MD-Paedigree, a Big Data and Decision Support tool for mining Europe’s first

social medical network on Paediatrics

News & Events

Related International Events

Table of Contents

SCANA Cloud-based Analytic Pipeline for Advance Cancer Prevention and Diagnostics

The identification of genes that are mutated and hence drive oncogenesis has been a central aim of cancer research since the advent of recombinant DNA technology. SCAN is a cloud computing-based pipeline approach, supported by multiple cloud computing infrastructures, to be run gene mutation detection software tools in “just enough, just on time” manner.

Wei Xing – CRUK Manchester Institute, University of Manchester, United Kingdom

4

Overview of SCAN Architecture

Today large-scale populations genome studies are urgently needed to address important issues of cancer treatment and personalized medicine. For studies on this scale, it takes large amount of patient samples to compare and contrast the patterns in order to assess whether the genome can tell us more about the undue burden. SCAN application is designed for such genomic and proteomic study. As a cancer detection pipeline, SCAN processes large amount of genomic and proteomic data in order to identify the driven mutation of tumor samples, and then to associate the identified mutation with protein functions within a cell signal network.

Taken advantages from CELAR, we design SCAN pipeline, supported by multiple cloud computing infrastructures, to be run in just enough, just on time manner. More precisely, CELAR can provision the resources required for each type of biological application during the stages of SCAN pipeline execution, and also CELAR is able to allocate the computing resources to SCAN pipeline runs according to the size of its genome data. The key objective of SCAN application is to match the resource demand required by a variety of bio-applications or by different volume of cancer data efficiently and economically. SCAN is comprised of a number of genomic and/or proteomic applications, which may incorporate multiple levels of biological information within their studies such as phenotype, genotype, expression profiling, proteomics, protein interaction, metabolic analysis and physiological measurements, etc.

As shown in Figure 1, SCAN is designed as a three-layered system in order to minimize the affect of the changes of cancer data or analytic tools: the bottom is a data layer to represent data. In the middle, it is data analytic layer to process data. On the top is the user interface. Generally SCAN pipeline has three key components: data entities, process services, and management or control engine.

SCAN is built for highly sensitive and accurate detection of sequence variants following reference genome mapping, infer protein pathway and signaling network, and identify the links between genotype and phenotype. SCAN is thus also designed as open module architecture to encapsulates various bio-applications or data analysis tools into SCAN as modular, i.e., SCAN Analytic Tool Boxes. Also each type of cancer data will have its corresponding SCAN Data Wrapper to hook-up into SCAN pipeline. The central data space for integrative data analyses is the SCAN databases.

5

SCAN: A Cloud-based Analytic Pipeline for Advance Cancer Prevention and Diagnostics

SCANUser Interface (CELAR cEclipse)

Computing services

Data services

SCAN Controller

Analytic Tool Box(Genome Process)

Analytic Tool Box(Proteome Process)

Analytic Tool Box(Image Process)

Analytic Tool Box(Integrative Process)

Data Wrappers SQL DBs(mysql, postgres)

no-SQL DBs(cassandra, hbase)

Next GenerationSequence Data

Drug Data Molecular Data Clinical Data Image DataPublic Databases

Figure 1. Overview of SCAN Architecture

SCAN Key Components

Running SCAN on Top of CELAR Platform

SCAN is described and submitted to underneath cloud infrastructures by using CELAR. CELAR provides an application template to describe applications. The template includes a service topology consisting of the various node templates (application components), relationships, elasticity policies and management operations, based on the OASIS TOSCA specification.

SCAN as an open module application system that can run on top of multiple cloud-computing infrastructures within CELAR framework. In particular CELAR allows SCAN support computational and data elasticity so that CELAR can intelligently orchestrate and adjust the computing resource allocation according to needs of cancer diagnose and the nature of cancer data of individual patients.

Analytic Tool Boxes (ATB): It is designed as individual computing service, which process and analyzes genomic data, proteomic data, or image data.

Data Wrappers (DWAS): DWAS is used to retrieve cancer data and feed into the ATB. SCAN has to deal with •heterogeneous biological data. Those data can be any kind of formats, storage in different storage systems.

•Databases : SCAN employs database systems as the means to manipulate heterogeneous cancer data, •including public human gene reference, protein sequence, gene function etc.

•SCAN Controller : The controller will be the main orchestrator to talk to CELAR middleware components in •order to enable the elasticity and dynamic provisioning etc.

SCAN: A Cloud-based Analytic Pipeline for Advance Cancer Prevention and Diagnostics

Genome Data Processing

Linux HPC Clusters Windows & Linux Servers Web Servers Shared memory systems

Image Data Processing

Proteome Data Processing Integrative data analysees

CELAR

6

Conclusions

SCAN application will be submitted to CELAR platform using the Application Submission Tool (AST). AST will first parse the SCAN description, and then CELAR Manager will broker the required computational resource to the application.

Systematic studies of the cancer genome can provide us a global view of the molecular architecture of complex traits and are useful for the identification of genes, pathways and networks that underlie cancer. Because cancer cells have a large variety of relatively rare mutations, it places a computational challenge to process hundreds of millions of short DNA sequences (corresponding to billions of DNA nucleotides) in a cost-effective way.

We design an integrative cancer detection pipeline (SCAN) to run on top of CELAR platform. So that SCAN can be supported by elastic resource provision of any underneath cloud infrastructures in a transparent and customizable manner. It enables performance and cost of SCAN execution remain within an objective function specified by the SCAN users, such as cancer researchers or clinical doctors.

CELAR TOSCA Service Topology For SCAN Genome Data Process

Data Wrapper

Analytic Tool Box(bwa)

Analytic Tool Box(GATK)

Database(MySQL)

Connects to

Node Template Relationship Template

Connects to

Connects to

Data Wrapper Database (MySQL)

Analytic Tool Box(bwa)

Data Wrapper

Data Wrapper

Analytic Tool Box(GATK)

Analytic Tool Box(GATK)

Analytic Tool Box(bwa)

Analytic Tool Box(bwa)

Analytic Tool Box(bwa)

7

ClouT - Cloud of Things for empowering the citizen clout in smart cities

Urban regions around the world are aiming to offer a more efficient, sustainable, and quality life for their citizens. ICT plays a substantial role in achieving these aims and thus in meeting the growing demand for smarter and more efficient cities.

Isabel Matranga, Engineering Ingegneria Informatica SpA, Italy

8

Living with the Internet of Things

A project with ClouT

Healthcare and assistance for the elderly

ClouT in action: aged care

By literally allowing anything to be interconnected, the Internet of Things (IoT) has great potential to increase the smartness of cities and city life. IoT devices can capture data not only about temperature, air quality and movement, but also about user preferences, intentions, medical conditions and so on. This wealth of data can inform cities, offering a wide range of options for facing emerging challenges such as efficient energy management, economic growth, development and citizens’ wellbeing and health.

It is envisaged that in the coming years the IoT will create tens of millions of new objects and sensors, all generating real-time data. The sheer volume of data coming from the IoT will require big-data and storage technologies, and this is where IoT meets cloud computing: cloud technologies may provide the computing power required to process the data coming from billions of IoT devices.

The ClouT project brings to the stage of smart cities the two sets of ICT technologies: IoT and Cloud Computing. ClouT sees a group of European and Japanese organizations coming together to cooperate in building tools and applications for cities and their citizens.

Prototypes of the ClouT solution will be deployed in four cities: Santander and Genoa in Europe, and Mitaka and Fujisawa in Japan . Each prototype city will then develop innovative applications useful in domains such as security, transport, healthcare and entertainment.

The Japanese city of Mitaka, for example, will focus providing on healthcare and assistance to the elderly. In Japan, as in Europe, an important challenge for local governments is improving quality of life for the elderly. In recent years the number of elderly people living alone in Mitaka has increased, along with the need for an environment that supports them in their everyday life. The city feels the need to generate a “symbiotic relationship” between everyone – its citizens, the private sector and the public sector – aimed at supporting each other within the city environment.

In this context, Mitaka intends to evaluate the ClouT Platform in terms of its ability to collect and leverage human information. It will effectively match participatory sensing data (i.e. data provided by the citizens themselves) with open city data (i.e. static data which the local government holds). This data match and processing will support Mitaka’s local government and communities in responding to the needs of its elderly citizens.

Mitaka’s applications are classified into two types:

Using these applications via the ClouT platform, Mitaka aims to create an environment in which the elderly will not feel isolated at home anymore, and where they can receive support and a set of everyday suggested activities to help them to keep healthy.

ClouT - Cloud of Things for empowering the citizen clout in smart cities

The Interoperable Data Storing field trial plans to create and operate a Dynamic Town Data Storage which collects the data of city events and the citizens’ activities.

The Actuating-People field trial uses data from the Dynamic Town Data Storage facility to generate customized city information aimed at motivating its citizens to be active and participate in city life.

9

Through the ClouT platform, the elderly can receive suggested walking routes – based on their health, past walking itineraries and weather conditions. They can receive information about coming events, based on their individual preferences and past activities, and even shopping support can be provided via suggestions on the best pedestrian routes and shop information.

By providing a safer city environment, Mitaka also aims to motivate elderly people to go out more and to have a healthier lifestyle.

For further information on ClouT project visit http://clout-project.eu/

10

A cloud-based PHR-EHR integrated environment for personalization of care

Social trends and ever-decreasing resources are pushing healthcare systems towards new, sustainable, patient-centred and multidisciplinary organizational models. As part of this, Electronic Health Records (EHR) are changing European healthcare systems, moving from a vertical approach towards solutions able to organize and manage entire healthcare processes based on integrated care.

Following this trend, Dedalus designed and commercialized X1.V1, an interoperability platform adopted by several public and private institutions to provide the infrastructure for a feasible EHR at local and regional levels.

Davide Guerri, Serena La Manna, Marco Lettere, Vincenzo Cestone, Sergio Di Bona, Dedalus S.p.A., Italy

11

Patient empowerment

X1.V1: personal health record + electronic health record

X1.V1 architecture

Introducing AmI

To improve personalized care, patients must be empowered to become more involved in their own care processes. Patient empowerment embraces the idea that everyone has the right to make their own choices about their health and care.

European “Patient Guidance Services” concept encourages the use of eHealth technologies to enable patients to actively participate in care processes and disease prevention, and Personal Health Records (PHR) aim to help patients to directly manage their own clinical data and information.

The X1.V1 platform is based on a service-oriented architecture and implements the XDS.b (Cross-Enterprises Documents Sharing) integration profile of the IHE (Integrating Healthcare Enterprise) standard, which natively incorporates a multi-repository/registry architecture. X1.V1 also implements a Master Patient Index (MPI) that guarantees the unique, conflict-less identification of patients, a Master Code Index (MCI) for terminology indexing, and an innovative component for Citizen-centred Workflow Tracking (CWT).

The X1.V1 Platform is deployed on a set of virtual machines, which can already be released in a private cloud Infrastructure as a Service (IaaS).

To realize the PHR-EHR integrated environment described above, Dedalus extended the X1.V1 platform and developed a cloud-based portal upon it. In particular, the X1.V1 architecture has been extended with a new component, the AmI module (name inspired by the Ambient Intelligence and Internet of Things paradigm).

The AmI module can collect raw and/or structured information from different devices and store it in a dedicated database, associating each piece of data with a specific patient stored in the MPI module. The AmI module also includes configuration facilities for extracting and summarizing the acquired data, and for producing documents according to the standard CDA2 (Clinical Document Architecture) PHMR (Personal Healthcare Monitoring Report), and storing them in the repository of the X1.V1 platform.

The Dedalus portal exploits the deployment of X1.V1 in a private cloud and is delivered as Software as a Service (SaaS), providing a set of standard-based and open services for patient identification and data management.

X1.V1 is a cloud based PHR-EHR integrated environment that supports patient-oriented short closed-loop care services, involving:

Patients, who have the right to activate their PHR via a cloud-based portal. Each PHR is a specific configuration of the X1.V1 platform, accessible only by the patient, or, with the consent of the patient, the GP. Information provided can be integrated with an EHR system.

Prescriptors (e.g. GPs or primary care doctors), who can prescribe therapies and/or the activation of personalized monitoring devices.

Dispensers (e.g. pharmacists), who can provide drugs, configure devices and activate, with the consensus of the patient, telemonitoring services.Patients can choose to apply devices and to send monitoring or therapy data flows directly to their PHR. GPs can monitor data provided by the patient. If necessary, other healthcare operators can access patient data through their EHR, if integrated with the patient’s PHR.

12

A cloud-based PHR-EHR integrated environment for personalization of care

The first time a patient logs into the Dedalus portal, he should create his own user profile and link different care services to it, choosing from a list of the telemonitoring systems that can be associated with the Dedalus PHR. Prescriptors can then prescribe telemonitoring and other services; dispensers can create a new account for the patient in the selected telemonitoring system, and then associate any devices to the PHR of the patient.

The credentials adopted in the two sites are associated through a new key (user key), which is generated according to the OAuth 2.0 specifications (http://oauth.net/2), and all the information exchanged will be linked to the identified patient. In this way, information acquired by the telemonitoring device is stored in the AmI database and becomes available and accessible in the patient’s PHR. Moreover, if the patient agrees, the data can be accessed by a GP and forwarded to the EHR of the relevant healthcare system(s).

To test and verify the overall architecture in a real test case, Dedalus collaborated with iHealth (www.ihealthlabs.com), a company that designs innovative, mobile personal healthcare products for self-monitoring.

Extended versions of the AmI module are now being studied with the aim of integrating a DSS/expert system to create documents from raw data using automatic or semi-automatic algorithms, based on configurable rules.

The SaaS-based architecture of the Dedalus PHR allows device vendors to join this care model, by guaranteeing that device-generated information can be shared across the healthcare systems, with GPs, specialists, hospitals and so on.

With the patient-centric closed loop, and the openness of the healthcare system, this environment will support continuity of care and the delivery of personalized care.

Security of healthcare data is a prime concern. As such, Dedalus adopted a three-layered approach that involves authentication of the client (the AmI component), the user (the patient) and the services in the contract (the monitoring devices).

Secure linking

iHealth test case

13

A cloud-based PHR-EHR integrated environment for personalization of care

EHRhealtcare operators

GP

Citizen

TelemonitoringSystem

PHRPORTAL

Amlcomponent

NOTIFICATIONMANAGEMENT

ORDERMANAGEMENT

MASTER PATIENTINDEX

REGISTRY DOCUMENT REPOSITORY

MASTER CODEINDEX

SINGLE SIGN ONENTERPRISESERVICE BUS

NOTIFICATIONMANAGEMENT

ORDERMANAGEMENT

MASTER PATIENTINDEX

REGISTRY DOCUMENT REPOSITORY

MASTER CODEINDEX

SINGLE SIGN ONENTERPRISESERVICE BUS

AmlDB

PHMR

DOCS

X1V1

X1V1

Industries across the world are undergoing a transition to a fully digital environment. New technologies are producing spirally amounts of data which must be reliably and economically maintained for decades.

The ENSURE project has created a reference implementation for a cloud based, long-term digital preservation system. ENSURE works with industry specific cost models to help dynamically configure a system which meets all legal and business requirements while minimizing preservation costs. To maintain the data, the project has developed a preservation-aware storage system called PDS which can run on top of OpenStack’s Swift object storage. PDS includes a Storlet Engine that allows for units of computation called storlets to be executed on the storage server, saving the costs and latencies. A variety of storlets have been developed as a reference. The Storlet Engine supports distributed storlets - where similar to the Map-Reduce framework, operations on large data sets can be parallelized across multiple storage nodes and then the results combined.

Orit Edelstein, Simona Rabinovici-Cohen, Eliot Salant – IBM Research Haifa, Israel Wei Wang – Philips Digital Pathology Solutions

14

ENSURELong term preservation of digital data with benefits

The ENSURE system’s architecture consists of:

ENSURE Architecture

PDS and Storlet Engine

Digital Pathology Storlets – demonstrating the power of ENSURE storlets

15

A set of plug-ins that provide specific functionality such as format management, regulatory compliance, integrity checks, and access to specific storage clouds.

A runtime Service-Oriented Architecture (SOA) framework that allows an OAIS (standard long term digital preservation functional model) solution to be created from those plug-ins needed to meet a user’s requirements, including any economic considerations.

A Configurator and an Optimiser which use cost/quality analysis engines to create and evaluate a proposed preservation solution.

An Image Storlet which reduces the bandwidth requirement in transferring images by only retrieving the relevant pixel data from the preserved images.

ENSURE is structured in two layers: the Configuration Layer and the System Runtime. The ENSURE Configuration Layer runs prior to the initial deployment of the preservation solution and re-runs periodically, in particular if there are major environmental changes. It suggests potential preservation solutions for selection; it optimizes the suggested solutions based on cost and quality estimations.

The ENSURE System Runtime is the SOA infrastructure for executing the plug-ins selected by the Configuration layer. This layer provides data management and archival storage services, as well as ingest and access services. It interacts with external storage services which provide the physical space for storing the preserved data and watches for environmental changes that may require the system to be reconfigured.

Preservation DataStores (PDS), the storage infrastructure of ENSURE, is preservation-aware and offers storage-side computation which has the ability to extract the maximum value from stored data. This is achieved via a consolidated storage platform for objects and computational processes (storlets) that are triggered and subsequently executed close to the data. The storage platform utilizes OpenStack Swift open source for cloud object storage and adds to it a Storlet Engine. This transforms the traditional archival storage to a richer service with automated preservation processes and potentially higher business value.

The innovative Storlet Engine allows data processing within the storage thus enabling preservation functions to be offloaded to the storage via storlets close to the data. This enables adding extensions to the cloud storage and creating workload-based solutions without changing the storage cloud internal code. Furthermore, the Storlet Engine vastly reduces bandwidth consumption, enhances security, saves costs of client infrastructure and supports compliance, enabling better cost-efficient preservation models.

The Storlet Engine provides special service storlets that can be used either by external clients or called by other running storlets to ease storlets development. One such service storlet is the Distributed Storlet, a compound storlet that executes multiple other storlets in parallel, and merges their results. This storlet is intended for analytics, where distributed data-intensive processing on multiple objects is required.

Digital Pathology for tissue imaging results in very large data files. Once stored in the preservation system, the performance of retrieving and analyzing such images can be greatly improved by the Storlet Engine because of its “processing close to data” feature. ENSURE has designed a number of specialized storlets to highlight the advantages of this technology such as:

ENSURE – Long term preservation of digital data with benefits

ENSURE – Long term preservation of digital data with benefits

An Image Alignment Storlet which aligns related images in the preservation system, so that matching regions are returned from these images when requested.

A Cell Detection Storlet, which detects cell pixels and demonstrates how the Distributed Storlet Engine can be used to improve the performance of image analysis on preserved images through parallel processing. Using distributed storlets, ENSURE can efficiently and rapidly analyze digital pathology images in the range of 20Gb running on even low end, commodity servers.

16

Let’s talk about new services for patients and healthcare organizations, not cloud computing

Healthcare services are driving the digital health innovation. We can now talk about digital health 3.0, thanks to healthcare integration with advanced IT services, such as cloud computing.

Maria Beatrice Fasano, ConnexxaLife, Italy

17

Until a few years ago, most healthcare organizations focused their IT interest almost totally on enterprise resource planning (ERP) systems for management and accounting. Now attention has shifted to their core business: clinical activity.

Driving this change is the pervasiveness of IT in patient care. At the same time, demand for e-health applications has grown, linked to the need to access and share data and information anywhere, and any time, in order to have personalized and continuous patient monitoring. As a direct result, there has been a steady increase in services provided and data to be managed.

As a consequence, the IT infrastructure in hospitals – both public and private – has become more and more complex, dealing with:

Driven by these needs, healthcare organizations are taking the first steps towards cloud computing. There has been an overall increase in spending for cloud adoption, targeted at improving efficiency, but above all (hopefully!), at providing improved service and innovation to patients. It is evident that cloud computing can accelerate the adoption of new technologies, improving the delivery of existing services and the proposition of new ones.

In this light, Connexxalife, an innovative Italian software company focused on developing cloud and mobile healthcare applications, has developed and delivered a high-level solution, called Galileo iClinic. Successfully adopted by several Italian hospitals, including the Hospital Papa Giovanni XXIII in Bergamo, Italy, Galileo iClinic ensures that:

Thus to support healthcare facilities, our information systems must meet high requirements for safety, reliability and usability.

Changes in healthcare culture

Turning to cloud

Galileo iClinic in action

18

Let’s talk about new services for patients and healthcare organizations, not cloud computing

the increasing cost of managing and maintaining IT infrastructure, and

the need for interoperability between heterogeneous software and information.

Accessing medical records is faster and more intuitive

Doctors consult with each other remotely and in real-time

Treatment plans are updated rapidly, eliminating errors

Images such as X-rays can be shared easily with patients

Encryption means that medical files are secure

Solutions are flexible enough to work inside and outside the hospital

Galileo iClinic is suitable for big organizations as well as small and medium-sized healthcare organizations. In fact, cloud computing offers small and medium-sized users a great opportunity, because they can use and take advantage of IT services that until now were used only by large organizations.

Increasing need for IT integratio

n

Transactionprocessing

systems

Knowledgebased

systems

Informationsystems

Nee

d fo

r rat

iona

l dec

isio

n m

akin

g pr

oces

s

Stan

dalo

ne a

pplic

atio

ns

Inte

rope

rabl

e ap

plic

atio

ns

Inte

grat

ed a

pplic

atio

ns

Automation

Let’s talk about new services for patients and healthcare organizations, not cloud computing

Cloud computing can critically impact many aspects of healthcare, including the implementation of mobile electronic patient records, telemedicine, digital services for the citizen, the dematerialization of clinical documents and computerized drug management.

We must talk to healthcare organizations in these terms: cloud computing is not just another technology to adopt; it is an enabling factor to implement new services for patients and improve existing ones.

This approach can be difficult for IT companies, as they are often tied to technical discussions. Therefore it is important to create knowledge about cloud computing. However, we must spread cloud in terms organizational benefits, rather than speaking of technicalities like bits, elasticity, and so on.

In this context, the role of political institutions (local, as well as European) is essential. Political institutions and their IT agencies must spread the vision of how a healthcare system can become efficient and truly patient-centric, and not focus solely on regulating procedures for protecting sensitive data, or indicating standards for communication and access to information.

Finally a few words about big data, a theme that also involves healthcare organizations, since they collect, manage and analyze clinical data.

Cloud computing is also an answer for this need. If we talk about the context of a single hospital, the direct result of a single patient’s data analysis is of course a deep understanding of the specificity of the disease of the single individual and, therefore, a targeted therapy.

But let’s try to imagine a technology that gives us the ability to collect and analyze real-time data: about patients suffering from chronic diseases, not necessarily hospitalized, with extended monitoring over time, and about genetic and environmental factors which may affect genetic disease. Let’s now imagine extending and accelerating that data collection, thanks to the same technology, collecting data from a large population, getting a number of examples more valid in term of statistics (compared to current medical research) and more quickly.

The consequent benefits for better patient care and the prevention of disease are evident. Cloud computing can be an enabler for this scenario, and so is vital to our health.

On their own terms…

Translating the vision

Big data: big benefits

UMTS3G

UMTS3G

Rete UMTSRete UMTS

CUP RIS/PACS MPI ADTLaboratoryAnalysis

Galileo iClinic

Hospital Network

Middleware HL7 / DICOM Gateway

Rete WIFI

FD - TS

Store images

19

PINCLOUD

Themistocleous M., Koumaditis K., and Vassilacopoulos G. - Digital Health Services Laboratory (DHSL), Department of Digital Systems, University of Piraeus, Greece

INTEGRATED E-HEALTH SERVICES OVER CLOUD

20

an aging population, with increased demand for specialized healthcare services (due to chronic diseases, for example),

the need for increased efficiency with limited financial resources (resulting in a reduced staff:bed ratio, for example),

increased demand for accessible care outside of hospitals (home care, for example).

Most developed countries face significant problems regarding the provision of healthcare services:

Advances in information and communication technologies have gone a long way towards tackling these problems. Nevertheless, homecare and healthcare services require intergration of e-health services.

E-health integration is commonly approached by adopting integrated patient-centred care. Integrated patient-centred care requires continuous, long-term coordination across professionals, facilities and support systems. This approach can be seen in national healthcare strategies that encourage patient involvement in their healthcare treatment. In the USA and Europe, online personal health records that allow patients to manage their health data have also emerged.

These patient-centred approaches empower patients, offering a visual overview of their course of treatment, allowing them to take their own measurements, and to provide verbal and written inputs. In terms of technology, this empowerment is enabled through information-sharing.

Many of these applications are based on service-oriented architecture (SOA) and cloud computing: state-of-the-art technologies that can be used to provide efficient, scalable, portable, interoperable and integrated IT infrastructures that are cost effective and maintainable.

Yet despite the significant importance of these technologies, the healthcare sector has not paid them much attention. For that reason, it is extremely important to integrate healthcare services. All the challenges currently facing the healthcare sector can be addressed through an innovative integrated e-health services platform that utilises advanced technologies like cloud computing and SOA.

Providing INtegrated e-health services for personalized medicine utilizing cloud infrastructure (PINCLOUD) is a proposed multidisciplinary project that combines cloud computing, SOA, homecare telemedicine technologies, e-personal health record (e-PHR), e-prescribing, e-referral and e-learning.

Integrated e-health and patient-centred care

Embracing state-of-the-art

The PINCLOUD project

21

PINCLOUD: INTEGRATED E-HEALTH SERVICES OVER CLOUD

Figure 1: PINCLOUD

(a) dynamic scaling, scalability, elasticity, security, fault tolerance, accounting granularity, cost allocation, and interoperability in clouds, (b) SOA governance, SOA architectural design, QoS and security, (c) resource allocation and management in e-learning systems that run over clouds, (d) the development of remote homecare telemedicine applications, (e) the implementation of ePHR, e-referral and e-prescribing applications, (f ) the integration and management of all the aforementioned applications, and (g) the management, governance and security issues related to the integrated e-health environment.

PINCLOUD seeks to integrate different application components, leading to the provision of an ‘anytime, anywhere’ end-to-end personalized disease monitoring and medical data service that ensures independent living, regardless of age. Constant monitoring can enhance early detection of emergency conditions and diseases for at-risk patients, and also provide a wide range of healthcare services for people in need.

Thus we can use the power and capabilities of cloud computing (e.g. dynamic scaling) to overcome the limitations of currently deployed healthcare applications. The proposed applications will be developed using SOA and web services.

Systems integration is always a challenge, as multiple users and organisations with different views, requirements, policies, interests and demands interact. For this reason, issues related to SOA governance, security, quality of service (QoS), architectural design and resuability are considered important, and further research is required to address these issues.

PINCLOUD will deliver significant benefits to users, society, economy, and academia, and it will extend the body of knowledge through attempts to investigate related issues, such as:

Beyond the state-of-the-art

Related investigations

22

PINCLOUD: INTEGRATED E-HEALTH SERVICES OVER CLOUD

H

Doctor’s O�ce

Health Insurance

Hospital

Home Care

Cloud Computing

PharmacyDiagnostic Center

Web

The advances proposed in PINCLOUD involve the integration of PHR systems with pervasive health monitoring systems, as well as user-friendly access to e-referral and e-prescription services provided via a cloud computing platform. To this end, we hope to develop an innovative, comprehensive and robust platform that will address the challenges facing our healthcare systems.

23

PINCLOUD: INTEGRATED E-HEALTH SERVICES OVER CLOUD

SOFTCARE:

Francesco D’Andria, ATOS Barcelona, Spain and Elisabetta Di Nitto, Politecnico di Milano, Italy

Applying the principles of social inclusion to adults with mental health problems is increasingly seen as desirable. ICT can offer support to this problem by providing E-Health systems that support careers in their daily activities by continuously monitoring such patients. Recent figures, however, have shown that the usage and adoption of E-Health in the emerging countries is very low. This is probably is due to economic factors. In fact, most of the time proprietary E-Health systems are quite expensive, and require a long-term maintenance. The software as a service paradigm, supported by the cloud, may relief this problem as it can result in a reduction of the fixed costs typically deriving from the adoption of such systems. In this paper we show how, relying on the SeaClouds multi-cloud platform, it is possible to develop such kinds of systems.

Multi cloud-enabled platform built around the needs of elderly people

24

SeaClouds [2] works towards giving software providers the capability of “Agility After Deployment” for cloud-based applications, by supporting developers and complex application managers through the creation of an open source platform that leverages open standards (such as OASIS CAMP[3] and TOSCA[4]) in order to support the deployment of applications over multiple-clouds, the monitoring of such deployments, and the migration of application modules across different (both public and private) cloud providers if needed.

SeaClouds aims at homogenizing the management of different and heterogeneous cloud providers and at supporting sound and scalable orchestrations of complex software systems across them. Systems developed with SeaClouds will inherently support the evolution of their constituent services, so as to easily cope up with needed changes, even at runtime.

Current cloud technologies suffer from a lack of standardization that prevents their adoption [1]. This issue is exacerbated in the case we need to deploy and manage complex and heterogeneous E-Health and Social Care systems as they typically run on multiple public/private infrastructures.

Nowadays, in fact, the adaptive management of complex applications deployed across multiple heterogeneous Cloud platforms is one of the problems that have emerged with the cloud revolution.

The recently started EU research project SeaClouds aims at providing seamless adaptive multi-cloud management of complex applications by supporting the distribution, monitoring and migration of application modules over multiple heterogeneous Clouds platforms.

INTRODUCTION

THE SEACLOUDS APPROACH

SOFTCARE: Multi cloud-enabled platform built around the needs of elderly people

Fig. 1. SeaClouds’ architecture.

25

SOFTCARE: Multi cloud-enabled platform built around the needs of elderly people

In SeaClouds, ATOS aims at implementing and assessing a heterogeneous E-Health and Social Care system as a multi cloud-enabled platform built around the needs of elderly people affected by a degenerative disease proving the following tools:

Heterogeneous E-Health systems providers are increasingly discovering that they can benefit of the public cloud for the non-sensitive functions while they can employ on-premises infrastructures or private clouds element for operations that are more sensitive or when regulatory requirements exist for data handling and storage (i.e. medical data repositories, etc.).

Certainty in terms of intended benefits, while cost savings continues to be a priority, increased speed of innovation has risen to the top for more experienced organizations. In this context, SeaClouds offers the following advantages:

[1] M. Armbrust et al.: A view of cloud computing. Commun. ACM (2010)[2] SeaClouds Project http://www.seaclouds-project.eu/ [3] OASIS Cloud Application Management for Platforms (CAMP) https://www.oasis-open.org/committees/tc_home.php?wg_abbrev=camp [4] OASIS Topology and Orchestration Specification for Cloud Applications (TOSCA) https://www.oasis-open.org/committees/tc_home.php?wg_abbrev=tosca

SOFTCARE: CLOUD-BASED SOCIAL SUPPORT NETWORK APPLICATION

EXPECTED BENEFITS AND CONCLUSION

Peer to Peer tools to maintain a close link between the patients and public/private health systems. This link will be useful to daily monitor the status of the patient. Educational tools (e.g. electronic library, video, braining games, etc.) to facilitate personal motivation and to enhance the notion of self-management.An interactive platform for music therapy to improve behavioural and psychological symptoms Video-conferencing utilities for the provision of a more thorough clinical image of the patient to the medical expert.Platform that integrate Social Networks. Elderly people are keen to maintain contact with the different generations of their family.

Contrast & compare different and heterogeneous Clouds for application in a fragmented market of difficult-to-compare cloud solutions.Reduce operational overhead with multi-cloud application management featuring simple governance complex application through unified interfaces, dashboard monitoring, unified metrics and user-defined SLA policies between Cloud platforms.Help alleviate vendor lock-in and lower switching costs in an ecosystem of “cloud adapters”, that empower the developer to migrate and maintain complex systems between competing private and public Cloud environment.

26

Bottom-up: Towards Supporting Personalized Medicine in the CloudHarry Dimitropoulos1, Anna Gogolou1, Herald Kllapi2, Omiros Metaxas1, Lefteris Stamatogiannakis2, Eleni Zacharia1, Yannis Ioannidis1,2, and MD-Paedigree EU-Project 1 I.M.I.S., Research Center “Athena”, Athens, Greece2 National & Kapodistrian University of Athens, Greece

Healthcare is experiencing a data explosion, driven by the plethora of medical tests, as well as, recent technological advancements, such as real-time data streaming from new mobile applications and wearable devices. Furthermore, healthcare data reside autonomously in disparate and heterogeneous sources in different hospitals or clinical centers. By making smart use of healthcare data, it is possible to identify disease signatures and build statistical simulation models going one step further towards personalized diagnosis and treatment. Our research team is working on developing systems capable of supporting the vision of personalized medicine by unifying health data from external systems under a Knowledge Discovery (KDD) platform that takes advantages of cloud technology and data mining.

27

DATA

DATA

ANALYSIS

STATISTICS

HEALTH

27

A big-data revolution in healthcare is underway. From birth, a person accumulates healthcare data of multiple forms, including clinical records, medical images, genetic data, and data streams produced via mobile applications and wearable devices.

One application of this data deluge is the development of model-guided personalized medicine. In developing personalized medicine, bottom-up (evidence-oriented) analysis is of fundamental interest. Such analysis attempts to identify latent factors (or disease signatures) that can explain and predict similarities and variabilities in drug therapies and disease evolution.

Bottom-up analysis requires 1) data integration from heterogeneous sources and 2) scalable analytics.

To this end, our research team is developing a Knowledge Discovery and Data Mining (KDD) platform, called AITION, for biomedical knowledge discovery, feature selection, vertical integration, and semantic modeling under uncertainty. In addition, we utilize Athena Distributed Processing (ADP) middleware providing advanced scaling capabilities, as well as, federated data access that supports a versatile execution of distributed algorithms on ad-hoc clusters and clouds.

The DCV module provides advanced techniques for data validation and preprocessing; checks for inconsistencies, missing values and outliers; computes medical scores; performs attribute discretization, and more.

The clustering module is a set of advanced clustering and similarity analysis techniques aiming to identify latent factors (disease signatures) and to group homogeneous patients. Following a mixed membership approach, a patient is characterized by a specific distribution (allocation) on multiple, latent disease signatures. Patient similarity is then computed by comparing such allocations using several metrics.

The simulation module analyzes correlations between variables, discovered latent factors and groups to deliver accurate and reusable predictive statistical simulation models based on Graphical Probabilistic Models (GPMs).

It implements state-of-the-art algorithms for Bayesian Network (BN) Structure & Parameter Learning, Markov Blanket induction and feature selection, and real-time inference. The system offers a user-friendly interface to explore the graphical models, as shown in Figure 1.

In addition, ontologies and a-priori knowledge can be incorporated, supporting a top-down (model driven) process that complements bottom-up (evidence oriented) analysis, providing a rich ‘natural’ framework for semantic modeling under uncertainty, where a domain expert is able to ‘seed’ the learning algorithm with knowledge about the problem.

The AITION KDD platform consists of three modules:

Bottom-up, evidence-oriented analysis

The AITION KDD platform

Data Curation and Validation (DCV) Clustering, and Simulation.

28

Bottom-up: Towards Supporting Personalized Medicine in the Cloud

Figure 1. The real-time inference capabilities of AITION

Over the last five years, our group has been building Athena Distributed Processing (ADP), a system for distributed data processing on the cloud.

ADP works to harness the power of cloud computing by defining language abstractions to declaratively express complex computation, and by designing an architecture with clear separation into components with well-defined semantics.

ADP offers a high-level language called ADP Query Language, which is based on SQL enhanced with user-defined functions and a new syntax that makes them easy to use. Thirty years of database technology has shown that declarative languages are important because they offer data and platform independence.

The functionality of the system can be extended with new user-defined functions, which can be as complex as needed. We offer a rich library of user-defined functions to support the AITION platform, including data import (CSV, XML), statistics (Pearson correlation), and more.

Over the last five years, our group has been building Athena Distributed Processing (ADP), a system for distributed data processing on the cloud.

ADP works to harness the power of cloud computing by defining language abstractions to declaratively express complex computation, and by designing an architecture with clear separation into components with well-defined semantics.

ADP offers a high-level language called ADP Query Language, which is based on SQL enhanced with user-defined functions and a new syntax that makes them easy to use. Thirty years of database technology has shown that declarative languages are important because they offer data and platform independence.

The functionality of the system can be extended with new user-defined functions, which can be as complex as needed. We offer a rich library of user-defined functions to support the AITION platform, including data import (CSV, XML), statistics (Pearson correlation), and more.

Federated data access and scaling utilizing ADP Bottom-up: Towards Supporting Personalized Medicine in the Cloud

29

Quantitative Medical Imaging in the Cloud: Enabling VIGOR++ with 3DNetHarry Hatzakis, Sören Grimm, Biotronics3D Ltd, UNITED KINGDOM and Costis Kompis , Vodera Ltd, UNITED KINGDOM

Just as digitization is dramatically transforming the access, distribution and review of medical images, technical advances in medical imaging modalities and computational methods are enabling the development of quantitative imaging.

30

Quantitative imaging involves a shift in clinical diagnostics, from a subjective interpretation of medical images to their objective evaluation, providing a quantified evaluation of both morphological and functional pathology. The extraction of quantitative metrics from medical images has enabled new measurements and comparisons, enabling the tracking of disease evolution.

The benefits of quantitative imaging are undeniable and cover applications from routine clinical procedures to drug development. Multiple challenges, however, still hamper the advancement of quantitative imaging, both in the lab and the clinic.

Despite continuous growth in data volume and complexity produced by diagnosis imaging, Picture Archiving and Communication System (PACS) remained, until the end of last decade, a very lucrative ‘physical’ hardware business, where images were stored on local pre-defined capacity servers and visualized on dedicated ‘high-end’ workstations. It was only at the 2011 Radiological Society of North America (RSNA) meeting that major PACS vendors, and various newcomers, unveiled their cloud-based systems. Around the same time we started working on VIGOR++.

The solution came from Biotronics3D’s 3DNet, which is regarded as the first advanced online medical imaging service and community. It uses the concept of Platform as a Service (PaaS) via cloud computing and provides functionalities such as:

VIGOR++ is a research project focused on personalized gastrointestinal tract models to facilitate accurate detection and grading of Crohn’s disease – a gastrointestinal disorder affecting millions of people of all ages.

The vision is one of a personalized virtual gastrointestinal tract model that can be updated with just an MRI scan. Being sufficiently accurate without the need for colonoscopy, the model should provide a quick and painless way to closely track the development of a patient’s condition, and to alter treatment accordingly. Equally, the next generation of patients with genetic predisposition to gastrointestinal disorders will receive early monitoring and advice on preventative action. This level of patient convenience can be enabled through local clinics, with remote access to clinicians as necessary.

The technology to enable this vision derives from multi-scale information from patients, including laboratory, MRI, colonoscopy and histopathology data. For this to happen, VIGOR++ clinical partners (AMC and UCLH) and scientific partners (TU Delft, ETH Zürich, and Zuse Institute Berlin) needed to handle large data sets and perform image analysis, pattern recognition and visualization tasks. At the same time, requirements for good performance as well as security, scalability and usability were high.

Medical imaging in the cloud

A personalized model of Crohn’s disease

3DNet: advanced online medical imaging service

Quantitative Medical Imaging in the Cloud: Enabling VIGOR++ with 3DNet

The system does not move data over the network. Instead, it uses streaming technologies. Therefore, studies with more than 5,000 images can be available in seconds for review, even over lower internet connections of 2Mbps. 3Dnet is developed using the .NET framework, Microsoft Internet Information Server, Microsoft SQL, Silverlight, C#, and C/C++, but it is operating-system independent. It works with WinOS, MacOS, iOS, Android and others. The system’s components are depicted in Figure 1. At its core there is a visualization engine and an analytics engine. These engines incorporate proprietary algorithms for advanced 3D rendering and image processing, but third party algorithms can be added via an API. 3Dnet follows the IHE profiles and supports the DICOM 3.0 standard: CStore, C-Find, C-Move, Query/Retrieve and Send/Auto-send.

workflow, analytics and archiving; centralized image storage and metadata repository; built-in text, data and image analytics that exploit multi-threaded processors; interface with HIS/RIS systems; and APIs for third party tools and diagnostic reporting solutions.

31

Quantitative Medical Imaging in the Cloud: Enabling VIGOR++ with 3DNet

Figure 1 The components of the 3DNet system.

Figure 2 Typical configuration and data flow in 3DNet

The system can be integrated with any HIS/RIS via its powerful HL7 broker. Figure 2 depicts a typical configuration and data flow in 3DNet, showing the way the core system interfaces with the Gateway, HL7 Service in order to enable access via Portals or dedicated Diagnostic Stations.

UserFront-end

ExternalInnovators

ISVDB Manager

Data Foundation Platform

Work�ow Platform

Visualization Engine

Analytics Engine

Management Engine

Metadata

Security& Audit

Manager

DataImport

ManagerMulti-tier storage

Web

Ser

vice

s

Dev

elop

men

t Pla

tfor

mEn

viro

nmen

t

DataGateway

32

Quantitative Medical Imaging in the Cloud: Enabling VIGOR++ with 3DNet

Success in clinical practice

Showcasing the benefits of cloud

Future challenges

Over 500,000 case studies have been handled by 3DNet system since Apr 2012 and each VIGOR++ case study consists of approximately 8,000 DICOM images. Related to gastrointestinal data, 3Dnet supports the following rendering modes: MPR, VR and VCPR (dissected colon). In the context of VIGOR++, we integrated novel automatic segmentation of inner and outer bowel wall segmentation (Figure 2, left), DCE registration, super pixel classification (Figure 2, right) and centerline extraction as well as measurement and annotation tools. The combined interactive visualizations have been supporting clinicians with the grading of Crohn’s disease severity, an important step towards determining treatment strategies and quantifying the response to treatment.

As demonstrated by the growing prevalence of cloud services, and through initiatives such as VIGOR++, cloud models can significantly contribute to quantitative imaging research and development.

Despite reservations about the technology and the requirement for a proper network infrastructure with high-speed bandwidth and more guarantees of security and data protection, the cloud has shown its potential to bring radical changes to medical imaging collaborative models, process automation, and workflow streamlining.

Cloud architectures such as that pioneered by 3DNet enable the creation of federated DICOM-based networks, located over distinct institutions, as well as the flexible, simple and secure provision of unique and integrated viewing tools, and management of resources – data, analytical applications and infrastructure. Such platforms improve recognition of the benefits of cloud models in research as well as in data analysis tool development, validation and collaborative use.

The transfer of methodologies from the lab to clinical routine or drug development still presents open challenges. New business models will be urgently needed to deal with the cost of applying these new approaches and optimizing their benefits.

Figure 3 Inner/Outer bowel wall segmentation integrated into 3Dnet (left) and Super pixel series overlaid onto scan data (right).

33

34

MD-Paedigree:a Big Data and Decision Support tool for mining Europe’s first social medical network on PaediatricsDr. David Manset CEO MAAT/GNUBILA, France and Prof. Patrick Ruch, HES-SO, HEG/University of Applied Sciences, Geneva, Switzerland

MD-Paedigree, a Big Data and Decision Support tool for mining Europe’s first social medical network on Paediatrics

Why Big Data is preparing medicine’s next quantum leap?

How?

Variety

From veracity to clinical decision-support

Beyond statistics: clinical decision-making as social networks

Simply because big data is preparing the next technological leap and technology is today’s main factor of social changes. Like medical imaging in the 80s, big data is about to re-organize medical practice.

The first and by far the largest factor of change is driven by biologists with the long tail of omics disciplines they are creating ex nihilo: genomics, proteomics, metabolomics (analysis interaction of small molecules), metagenomics (analysis of bacterial and organisms found in a human body) to cite a few. Being able to associate omics with clinical features will allow to build powerful risk assessment models able to associate a genetic profile, a behavior and a risk… provided that the data needed to generate these associations are made available one way or another. The recently launched MD-Paedigree project, co-funded with about 12 million Euro by the European Union, aims precisely at removing the numerous barriers which threaten the development of Big Data for healthcare. Challenges are numerous, technical on the one hand, societal and epistemological on the second hand. The societal dimension of the problem are relatively well-known and not all medical practitioners are digital natives, but let’s mainly focus our attention on technical and epistemological issues. Technical challenges are expressed by the 5 V: Volume, Velocity, Variety, Veracity, and Value. While Volume and Velocity represent obvious challenges with the exponential growth of mainly sequence data, we argue that the main challenges lie in Variety, Veracity (lack thereof ) and Value (hard to extract).

Indeed making sense of gene and gene products sequences demand the integration of extremely multimodal data (time-series, images, narratives, sequences …). The ability to make sense of these heterogeneous data interoperable is today an open scientific problem. The needed glue is mainly made of semantically rich resources, so-called ontologies, able to organize the knowledge of a given field so that machines can perform analytical operations on it. Today, virtually all serious development in the biomedical realm make intensive use of them.

While evidence-based medicine is constructed on a hypothesis-verification methodology, i.e. all patient having certain symptoms will receive a diagnosis following a step-wise process described in a clinical practice guideline, with Big Data clinicians will generate diagnosis and care plans based on a virtually indefinite set of evidences. For the first time since Descartes, the status of evidence is thus radically evolving. While evidence was traditionally supported by a crystal clear discourse (this drug is effective according to this proven mechanism of action), the new definition of evidence is: here are the data, from which the following (numerous) associations are derived. MD-Paedigree is thus exploring the development of a case base retrieval engine. For a given case in pediatrics, the idea is to retrieve files of children having similar anamnesis (e.g. age, gender, diagnosis, abnormalities, bacterial flora, genetic variants…) to identify the healthcare procedures likely to result in the best outcome. Unlike in evidence-based medicine, patients are not treated following a one-size-fit-all procedure, but instead the treatment is personalised to optimally suit the patient.

The need to federate European-wide clinical repositories to augment the project’s knowledge base is one of the challenges tackled by MD-Paedigree. Furthermore, MD-Paedigree attempts to balance the isolation of clinicians, whose expertise is questioned in the new paradigm: how can I build a crystal clear discourse out of tens of thousands of individually weak – yet statistically significant – evidences? MD-Paedigree’s answer is to offer the professional a tool for consolidating his/her decision by having recourse to a second opinion: all information pertaining to the case, as well as all evidence available in the knowledge base, can be shared with a colleague. Other pieces of the puzzle include the future exploitation of social media contents, with currently two promising threads: social networks for professionals vs. networks for patients. The latter is represented by private players such the US-based Sermo company, which connects more than 200,000 physicians. Sermo’s end-users can receives

35

recommendations from peers using fully anonymous channels. On the patient side, the social network trend is well illustrated by a recently launched IMI (innovative Medicine Initiative). The selected project should explore the use of Twitter to identify adverse drug reactions for sake of drug post-market surveillance.

MD-Paedigree, a Big Data and Decision Support tool for mining Europe’s first social medical network on Paediatrics

36

36

NEWS & EVENTS

Announcing the SUCRE EU-JAPAN Workshop – Research on Clouds and IoT in Europe and Japan: current status and ways to collaborateThe event, jointly organised by the SUCRE, OCEAN & ClouT projects , will take place at the EC premises in Brussels on May 16th 2014 and aims at contributing to the EU-Japan dialogue on open and interoperable Clouds. In this light, presentations and discussions will involve stakeholders and key industrial players from both regions. The workshop will also facilitate and foster the knowledge exchange between these two regions, compare success stories and pave the way to future cooperation opportunities between academia and Industry and between industrial players and policy makers from EU and Japan.Further information at www.sucreproject.eu

The SUCRE Healthcare workshopThe SUCRE healthcare workshop will take place within the eHealth Forum event in Athens, Greece on the 13th May 2014 in the morning. In this context, the SUCRE project will support the discussion about the use and facilitate the uptake of open-source development model solutions in cloud computing protocols. In particular, the consortium will examine how the interweaving of open source and cloud technologies can be further stimulated and adopted in the key sector of healthcare industry.The meeting will bring together practitioners from the industry, users and researchers to present recent results and discuss about the aforementioned topics. Further information and registration please visit the SUCRE portal.

CloudSource Magazine – Call for Abstracts Issue 4The Editorial Board of the SUCRE CloudSource Magazine is pleased to inform you that it is now accepting contributions for the 4th and last issue of the aforementioned Magazine to be published next September 2014.Prospective authors are invited to submit an expression of interest by sending via e-mail and/or via electronic submission on the SUCRE portal a short abstract (approximately 80-100 words) reporting recent developments and future visions in the Cloud and Open Source areas and specifically addressing the theme of Research on Cloud Computing in Europe and Japan: current status and ways to collaborate. Contributions should describe original research and development activities not published before. Very much appreciated are articles received from Japanese, South-East Asian, Indian, and Chinese key Cloud actors. Submission deadline and further information at http://www.sucreproject.eu/content/cloudsource-magazine-%E2%80%93-call-abstracts-issue-4

CLOUD FOR EUROPE workshopCloud for Europe partners launch pre-commercial procurementThe Cloud for Europe project supports public sector cloud use as collaboration between public authorities and industry. The project will carry out a pre-commercial procurement for research and development on cloud computing services for public administrations. The purpose is to research and demonstrate solutions to overcome obstacles for adopting cloud computing by the public sector. The tender will be launched in July/August 2014. For more information about the tender and the related events, please visit www.cloudforeurope.eu.

PROSE PLATFORMThe PROSE project is promoting the uptake of Free/Libre/Open Source Software (FLOSS) within European projects, through a software forge that supports the software development cycle of European projects, available at http://opensourceprojects.eu. OpenSourceProjects.eu will support your development process throughout your development cycle, and will help sustain the software beyond the projects’ completion, increasing the impact and visibility of FP7/H2020 project results. The platform will provide you with the necessary tools to also open your results to outside contributions and help generate a meaningful community within the FP7/H2020 research ecosystem. To get started register for an account on Open Source Projects and create your project today.Open Source Projects is provided and actively maintained by the PROSE Coordination Action (http://ict-prose.eu). The PROSE consortium members will actively maintain and enhance the platform for at least the next 5 years and are committed to the success and uptake of the open source projects platform as the next generation software development hub for Horizon 2020.

37

Nettropolis launches cloud mobile apps for public transport companies in GermanyEC funded MobiCloud project makes city transport companies more efficientAt the Nettropolis user conference in October last year a preview version of Nettropolis’ Nettro®MCD application, being developed as part of the MobiCloud project, was presented to representatives of public transport companies. The application, which is being built in cooperation with the Karlsruhe Transport Company (VBK), captures operational incidents on the spot and promptly informs responsible field staff, who can then respond and take action if needed. Since this introduction several public transport companies expressed their interest in working with Nettropolis to enable their personnel to also maximise the potential use of smartphones in their work. Full article at http://www.mobicloudproject.eu/#!pr5-english/c28l

EOpen Source Projects (OSP) Europe, in association with the ICT PROSE project will be hosting a workshopEOpen Source Projects (OSP) Europe, in association with the ICT PROSE project will be hosting a workshop on “Open Forge Eco-system distributing European FLOSS” at the upcoming Solutions Linux, Libres & Open Source 2014 event in Paris on May 20th 2014. The workshop will include an overview of the OSP Europe Forge, the overview of a mobile app which assists in the selection of FLOSS licenses and a session on FLOSS Business Model(s). The workshop and the OSP Europe booth (Stand E14) will be part of the programme at Solutions Linux which is being hosted at CNIT Paris la Défense, Hall Marie Curie, with our workshop in the “Miro” conference room May 20th at 14:00pm. Link to PROSE blog post http://www.ict-prose.eu/2014/04/14/open-forge-eco-system-distributing-european-floss-workshop/

38

RELATED International EVENTS

2nd Annual CLOUD COMPUTING East 2014The CLOUD COMPUTING ASSOCIATION (CCA) and the DISTRIBUTED COMPUTING INDUSTRY ASSOCIATION (DCIA) are proud to present this event that will focus on two major sectors of the economy that are leading the way in adopting cloud-based IT solutions: GOVERNMENT and HEALTHCARE. The speaking faculty will be made up of over 50 thought-leaders who will bring broad industry knowledge, technological savvy, and strategic insight.The event will take place at the Doubletree by Hilton in Downtown Washington DC. - May 15-16, 2014. For further information please visit http://www.cloudcomputingassn.org/events/T1401/overview.html

International Supercomputing Conference (ISC) 2014ISC offers researchers, engineers, vendors and students various opportunities to participate in the upcoming conference and exhibition. The calls for papers, posters, HPC in Asia posters, tutorials, and Birds-of-a-Feather (BoF) sessions are now open.This key conference will take place in Leipzig, Germany, June 22nd – 26th 2014. Further information please visit http://www.isc-events.com/isc14/#

7th IEEE International Conference on Cloud Computing 2014The goal of Cloud Computing is to share resources among the cloud service consumers, cloud partners, and cloud vendors in the cloud value chain. The resource sharing at various levels results in various cloud offerings such as infrastructure cloud (e.g., hardware, IT infrastructure management), software cloud (e.g. SaaS focusing on middleware as a service, or traditional CRM as a service), application cloud (e.g., Application as a Service, UML modeling tools as a service, social network as a service), and business cloud (e.g., business process as a service). This conference will take place in Alaska, U.S., June 27th – July 2nd 2014. For further information please visit http://www.thecloudcomputing.org/2014/

Euro-Par 2014Euro-Par is an annual series of international conferences dedicated to the promotion and advancement of all aspects of parallel and distributed computing.Euro-Par provides a forum for the introduction, presentation and discussion of the latest scientific and technical advances, extending the frontier of both the state of the art and the state of the practice. In addition, this conference provide a platform for a number of accompanying, technical workshops. Thus, smaller and emerging communities can meet and develop more focused topics or as yet less established topics. Euro-Par 2014 will take place in Porto, Portugal, 25th- 29th August 2014. For further information please visit http://europar2014.dcc.fc.up.pt/

Public Sector Cloud World ForumThe Public Sector Cloud World Forum will be the only Cloud event to secure a comprehensive line up of senior government IT officials from around EMEA. Guest keynotes from the US Government also feature in the agenda, showcasing their experience in Cloud adoption. The event and all related conference sessions are free to attend for public sector end-users, ensuring a high profile audience of Public Sector officials in attendance. The event will take place at Hotel Palace Berlin, Germany, 22nd-23rd September 2014For further information please visit the event website http://publicsectorcloud.eu/

IEEE CloudNet 2014Cloud Networking has emerged as a promising direction for cost-efficient and reliable service delivery across data communication networks. The dynamic location of service facilities and the virtualization of hardware and software elements are stressing the communication network and protocols, especially when datacenters are interconnected through the Internet. Although the “computing” aspects of Cloud technologies have been largely investigated, lower attention has been devoted to the “networking” aspects.The 2014 3rd IEEE International Conference on Cloud Networking (IEEE CloudNet 2014), part of the IEEE Cloud Computing Initiative, precisely addresses these aspects.The event will take place in Luxemburg, 8th – 10th October 2014.For further information please visit http://www.ieee-cloudnet.org/2014/cfp.html