1 Taking imaging to the cloudWhat are the benefits? Erik R. Ranschaert, MD, Radiologist

2 Copyright E R Ranschaert, ECR 2016Disclosure No conflicts of interest Copyright E R Ranschaert, ECR 2016

3 Copyright E R Ranschaert, ECR 2016Sources Shrestha, R. Real world challenges and benefits of cloud technologies, SIIM webinar Jan. 28 (2016). https://siim.org/general/custom.asp?page=web16_real_world Patel, R. P. Cloud computing and virtualization technology in radiology. Clinical Radiology (2012), 67(11), 1095–1100. Copyright E R Ranschaert, ECR 2016

4 Current issues in Imaging InformaticsImaging overload Focus on value vs. volume System integration Reforming radiological report Image sharing Big Data is big buzz word, data doubling every 18 mths, data explosion world-wide exabites, 44x increase from 2009 by 2020 – Data explosion Technologies around big data and accessibility to technologies thanks to emerging cloud – based methodologies opportunity to transform HC Source: SIIM Webinar Jan 28, 2016: “Real world challenges and benefits of cloud technologies”

5 Copyright E R Ranschaert, ECR 2016Imaging on the cloud The cloud is a combination of Computation Software Data access Storage devices No end-user knowledge of physical location and configuration of the system delivering the services Cloud Computation Software Data access Storage devices Copyright E R Ranschaert, ECR 2016

6 Is the Cloud the big “savior”?Copyright E R Ranschaert, ECR 2016

7 The (public) Cloud in personal lifeHCP’s are using the cloud for personal & professional purposes (e.g. Facebook) Music, photos and films in the cloud (e.g. Spotify, iCloud, Netflix) Documents in the cloud (e.g. Evernote, Dropbox) Sharing of clinical cases & social networking through the cloud (WhatsApp)

8 Era of online information sharingPrivate persons & non-medical businesses have become accustomed to moving many types of information via the Internet. Music, photos, videos, and financial information Society finds this means of information exchange efficient and desirable. Acceptance of this type of exchange of confidential information indicates a reasonable level of trust by users of this technology. Reality: there is a tremendous level of progress in cloud technology for HC services The challenge: to ensure the highest level of security and confidentiality for Internet-shared health care data. Copyright E R Ranschaert, ECR 2016

9 What is “cloud computing”?“A model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort” Mell, P., Grance, T. The NIST Definition of Cloud Computing. NIST web site. Accessed November 9, 2015 Copyright E R Ranschaert, ECR 2016

10 Characteristics & types of cloud computing5 Key characteristics 3 Types of services On-demand self-service Ubiquitous network access Location-independent resource pooling Rapid elasticity Measured service SaaS – running specific application through a cloud PaaS – using a suite of applications, programming languages, user tools IaaS – relying on remote data storage networks Source: SIIM Webinar Jan 28, 2016: “Real world challenges and benefits of cloud technologies”

11 Types of services Source: SIIM Webinar Jan 28, 2016: “Real world challenges and benefits of cloud technologies”

12 Copyright E R Ranschaert, ECR 2016Formats of clouds Public cloud services Used by multiple entities Private Dedicated to particular organisation Hybrid Combination of P&P Community Shared by X organisations Copyright E R Ranschaert, ECR 2016

14 Any-ology Era of image sharingOther specialties have needs for both static & cine image storage & retrieval Might have added value for collaboration More unified management of data and images needed Yet not sufficiently addressed by current methodologies Radiology Dermatology Ophthalmology Oncology Cardiology Pathology Otorhinolaryngology Neurosurgery Gastro-enterology Other “ologies” are following fast – entering digital era of imaging, needs better more unified ways to manage the data Cloud-based enterprise storage for all –ologies Challenge in silo –model! Cloud has single point of access, promise of using virtualisation! Source: SIIM Webinar Jan 28, 2016: “Real world challenges and benefits of cloud technologies”

15 Storage capacity neededStoring requirements are gradually more demanding : Production of medical imaging will continue to increase New techniques are producing large data volumes Hybrid imaging: PET-CT, PET-MRI Other “-ologies” also want to store digital images New cloud-based technique for image recognition (deep learning) allows to recognise and analyse images based on specific data patterns Needs enormous data-input & storage capacity Needs more computer power Is key to development of artificial intelligence Other “ologies” are following fast – entering digital era of imaging, needs better more unified ways to manage the data Cloud-based enterprise storage for all –ologies Challenge in silo –model! Cloud has single point of access, promise of using virtualisation! Copyright E R Ranschaert, ECR 2016

16 Traditional image storage Cloud-based image storageCloud-based storage Traditional image storage Cloud-based image storage Site 1 Site 2 Site 3 … Site 1 Site 2 Site 3 … App 1 App 1 App 1 App 1 App 10 App 1 App 1 App 1 App 1 App 10 Cloud based storage platform “Siloed” architecture Intensive & expensive to manage Low utilization of HW & storage Vulnerable to failure & downtime Inefficient bandwidth Different for each site “Virtualized” grid architecture Shared & cost-effective storage capacity Adaptive, self-healing, self-managing High speed Infrastructure is managed independent from application Single point of access: EPR, Portals Elimination of redundancy Source: SIIM Webinar Jan 28, 2016: “Real world challenges and benefits of cloud technologies”

17 Image exchange - PaaS Institution B Specialty Hospital Vendor-neutral, cloud-based medical image & document exchange Radiologist Institution C Community Hospital MRI scanner PACS server Image sharing across trauma centres, cancer centres Radiologist Radiologist/Radiographer Institution A Community Hospital Physician office Primary care PaaS. The cloud computing environment is used to transfer images to different institutions and clinicians in physically separate locations.

18 Key capabilities of cloud servicesAs easy as a social network: discuss patient cases, send messages, invite users, join networks Secure communication: enabled with SSL, secure DICOM encryption, XDS-I, audit trails Vendor agnostic: uploading from any device: CD, file folder, DICOM-based device Easy download option: edit & send incoming studies to local DICOM devices Built-in intelligence: analytics dashboard to visualize trends, specialty mix, indications etc. Copyright E R Ranschaert, ECR 2016

19 Reasons for image exchangePatients are using the Web, mobile access to information Easy gathering of images from hospital Easy to obtain second opinions Patient empowerment Broader market Increasing collaboration and specialty-oriented hospitals Personalized medicine Telehealth Easy referral process to outside facilities Avoid redundancy Easy access via cloud to information Trauma patients: faster, treatment preparation Hospital as gateway to fluently organised patient-centric care Liquid data, elegant transfer of studies Copyright E R Ranschaert, ECR 2016

20 Image processing - SaaSCloud provider 3D visualization SaaS Data processing Rendering of 3D images Cloud Study DICOM images Diagnostic centre PACS server 3D imaging software 3D imaging software accessed via web browser or thin-client downloaded to PC Hospital Reading radiologist Laptop Reading radiologist Institution A SaaS. DICOM images are transferred to the cloud in order to perform advanced 3D post-processing. The radiologist accesses the three-dimensional visualization software from any computer and is able to view and manipulate the rendered images.

21 Advantages of the cloudAllows economy of scale, “cloud-onomics” Improves performance and quality of service Automated upgrades No obsolete software Enables enterprise imaging Workforce, collaborative working Redefine planning of diagnosis and treatment Workflow improvement Increases quality of care workflow = BEST argument! more robust workflow clinicians have access to images and can work efficiently from multiple locations w full access to complete dataset complete diagnostic tools from any location one of big promises of cloud!! Copyright E R Ranschaert, ECR 2016

22 Copyright E R Ranschaert, ECR 20161. Economy of scale Efficiency of the provided service increases as the number of services being delivered increases. Lower average unitary cost of the service the fixed costs of the service are shared over an increased number of provided services. Ribeiro, L. S., Rodrigues, R. P., Costa, C., & Oliveira, J. L. (2013). Enabling outsourcing XDS for imaging on the public cloud. Studies in Health Technology and Informatics, 192, 33–37. Copyright E R Ranschaert, ECR 2016

23 Copyright E R Ranschaert, ECR 2016Cloud-onomics Unused resources Unused resources Dr. Jie Liu, Microsoft Research Copyright E R Ranschaert, ECR 2016

24 Cloud-onomics Lower initial investment Minimize capital expendituresReduced total cost of ownership Virtualization Faster deployment Simplified management Seamless scalability Economics of scale Source: SIIM Webinar Jan 28, 2016: “Real world challenges and benefits of cloud technologies”

25 Current trends in radiology2. Improves performance Current trends in radiology Advantages of cloud Imaging data sets are growing Increasing use of software for workflow processing Increasing pre- and post-processing requirements Increasingly distributed environment Flexibility for IT depts Better anticipation of fluctuations in user load No upfront investments for additional HW/SW Rapid expansion Robust reliability Better performance for clinical users, focus on clinical work instead of IT Copyright E R Ranschaert, ECR 2016

26 Copyright E R Ranschaert, ECR 20163. Enterprise imaging Solution for the workforce problem Creation of platform for collaborative working across sites. Organizations can optimize resources across multiple settings, providing speedy access to available radiologists and helping to balance workload Such collaborative working can help to accelerate care, while maintaining quality control Real-time access to data can help remove unnecessary delays, reduce the administrative burden, and improve operational efficiency. workflow = BEST argument! more robust workflow clinicians have access to images and can work efficiently from multiple locations w full access to complete dataset complete diagnostic tools from any location one of big promises of cloud!! Copyright E R Ranschaert, ECR 2016

27 True enterprise imagingAdvanced visualisation technologies are not limited to specialist radiologists Oncologists, vascular surgeons, cardiologists, liver surgeons get datasets and SW at their disposal, e.g. for 3D imaging Access possible from any location Online availability for multidisciplinary conferencing Possible to redefine planning of procedure and treatment Copyright E R Ranschaert, ECR 2016

28 Copyright E R Ranschaert, ECR 2016Workflow Radiographer > resident > attending radiologist > clinician Seamless access from anywhere at any time > access to complete imaging data sets incl. priors, and relevant clinical information Availability of right set of diagnostic tools from any location, appropriate viewer for referring clinician Streamlined image availability at the emergency dept. Easy forwarding to other hospitals, timely preparation for emergencies Copyright E R Ranschaert, ECR 2016

29 Copyright E R Ranschaert, ECR 20164. Quality of care Traumatology Complex diseases Oncology Second Opinions Teleradiology Telemedicine Situations in which it is crucial to have all relevant medical data available Copyright E R Ranschaert, ECR 2016

30 Copyright E R Ranschaert, ECR 2016Quality of care Traumatology Important for trauma centres and primary stroke centres Crucial decisions need to be made quickly Unnecessary transfers to other hospital can be avoided Preparation of treatment; planning resources before patient arrival Avoid redundancy; in 60% of cases re-scans are needed Copyright E R Ranschaert, ECR 2016

31 Copyright E R Ranschaert, ECR 2016Quality of care Traumatology Complex diseases and oncology Patients are often referred to a more specialized / tertiary medical centre. Increasing number of HC providers is focusing on a more specific type of cancer treatment. Increased collaboration between providers to improve overall oncological care. Multidisciplinary teams from different hospitals need to communicate on a regular basis to make decisions regarding the most appropriate treatment of patients. Cross-enterprise availability of the complete medical records is quite essential. Copyright E R Ranschaert, ECR 2016

32 Copyright E R Ranschaert, ECR 2016Quality of care Traumatology Complex diseases and Oncology Second opinions An increasing number of patients uses the Internet to find information about their health or disease. In this context patients also would like to have a better insight in their medical record, or they want to be able to provide access to this information to other health care providers. Copyright E R Ranschaert, ECR 2016

33 Copyright E R Ranschaert, ECR 2016Quality of care Traumatology Complex diseases and oncology Second opinions Teleradiology and telemedicine For teleradiology companies it’s important to have access to the full medical record for providing optimal reading services. Sharing of work lists across institutions or with external parties creates new possibilities to improve workflow management and to increase the availability of subspecialty advice and expertise. Copyright E R Ranschaert, ECR 2016

34 Meaningful connectivityThe Cloud plays important role in fostering better models for fluent exchange of images and information Interoperability profiles published by IHE (such as XDS) can be used to establish secure connections between PACS, RIS and EMR systems Intelligent pre- and post processing of imaging data is available with direct integration in clinical workflow Connectivity becomes more meaningful when information is used for data mining, data processing, cloud-analytics capabilities Copyright E R Ranschaert, ECR 2016

35 Healthcare is evolvingCurrent hospital-centric model Collaborative care model Care pathways Solutions Interoperable Patient data consolidation and integration Open Patient centric Hospital centric Episodic > Departments Patients Proprietary Data silos Source: SIIM Webinar Jan 28, 2016: “Real world challenges and benefits of cloud technologies”

36 Security Concerns Business environment Data governanceAuthorised access only (ID management) Integrity and Availability Prevention of data breach Data governance Data location Protection of privacy & confidentiality Clarity about data ownership & control Business environment Ensuring interoperability Ensuring portability (ability to reuse personal data across interoperable applications) Insufficient reliability (commitment to SLA’s) How safe are outsourced data when using these services? Is the ownership of data clear? General perception is that benefits outweigh the risks 3 major concerns: security, data governance, business environment Need for robust and sophisticated security strategy Data governance commitment is the key General perception: the benefits outweigh the risks Copyright E R Ranschaert, ECR 2016

37 Storage in the cloud Requires a robust and sophisticated security procedure to safeguard professional secrecy and information privacy of the patient It is vital to know where the server on which the information is stored actually is located how the information is encrypted which legislation is applicable (country) to the processing of health data. E R Ranschaert, ECR 2016

38 New Data Protection RegulationThe Regulation still has to be voted on by the European Parliament in plenary during spring 2016 ( March/April). https://www.huntonprivacyblog.com/2015/12/17/the-eu-general-data-protection-regulation/ E R Ranschaert, ECR 2016

39 Copyright E R Ranschaert, ECR 2016Upcoming EU directive 2016 Move to 1 single regulation for EU, will replace current patchwork of national laws Easier for companies to operate across EU Strengthens the citizens’ rights Cloud provider becomes data processor that must protect information it handles and stores on behalf of data controller Data controllers will have to put policies and procedures in place Data controllers will have to demonstrate that they have taken appropriate security measures to protect personal data of customers/patients Data subject Data controller Data processor Copyright E R Ranschaert, ECR 2016

40 Summary: reasons for cloudingEconomics Efficiency Patient care Takes IT from capital expenditure to operating expenditure Rapid data retrieval Reducing re-scan rate by 10-15% Cost control: closer match between cash flow and total system cost Efficient image-sharing Speedy transfer of data, fast treatment decisions Subscription and pay-per-usage models, flexibility and elasticity, on-demand expansion Streamlined workflow Avoid unnecessary patient transport No double examination costs Integrated intelligent pre- and post-processing Preparation of treatment before patient arrival Access to images, priors, clinical information at the point of care SAVINGS! MORE INFORMATION! QUALITY of CARE! Copyright E R Ranschaert, ECR 2016

41 Copyright E R Ranschaert, ECR 2016Example Recent agreement between Philips and Amazon Web Services (AWS) Delivery of back-up service for hospitals in Cloud Improved protection of data Increased need for storage capacity due to complex imaging techniques Matter of days to move all data to cloud Copyright E R Ranschaert, ECR 2016

42 Copyright E R Ranschaert, ECR 2016

43 Copyright E R Ranschaert, ECR 2016Let’s connect and collaborate Thank you! Copyright E R Ranschaert, ECR 2016