Intel next-generation-medical-imaging-data-and-analytics

Driving the Business Need for
Scale and Cost Containment
A perfect storm of business challenges,
emerging opportunities, and rapid
technology advances is reshaping
the medical imaging data landscape.
Together, these changes are making
cloud computing an increasing
necessity—and a critical opportunity—
for hospitals, clinics, radiology practices,
and other healthcare enterprises.
In the challenges column, value-based
payment models, the increasing global
demand for healthcare services,
Cloud computing and non-volatile memory
solid-state drives are spurring a new era of
medical imaging data analytics
Executive Summary
The rapid growth of medical imaging data and the rising need for collaboration
across the care team are making cloud computing an increasing necessity for many
healthcare enterprises. At the same time, increases in data analytics make imaging
data more valuable as a tool to drive research and improve patient care.
Intel and Carestream Health are collaborating to ensure practical, high-
performance solutions for picture archiving and communications systems
(PACS) and radiology information systems (RIS) workflows, in both on-premises
infrastructure and secure external clouds.
In a recent collaboration, Carestream Health demonstrated that new Intel® Solid-
State Drive (SSD) Data Center (DC) Family for PCIe® P3700 featuring Non-Volatile
Memory Express™ (NVMe™) provided a dramatic threefold increase in throughput
for a critical portion of the Carestream Vue™ workflow.1
This increased throughput
can provide a more responsive experience and greater productivity for clinicians.
By taking advantage of Carestream Health’s leadership in cloud computing for
PACS/RIS, including its broad use of Intel® technologies and its rapid adoption of
Intel innovations, health systems can reduce operations costs and manage data
growth while preparing for a new era of medical imaging data analytics.
changing regulatory climates, and the
emergence of the empowered patient
are creating a business imperative
to deliver higher-quality care more
efficiently and at lower cost. These
trends raise the need for healthcare
providers to securely and quickly share
patient health information, including
medical imaging data. Doing so can
improve collaborative care, enhance
patient satisfaction, and lower
costs by reducing duplicate tests,
procedures, and diagnostic studies.
Increasingly, it is a necessity to meet
regulatory requirements.
Ishai Tal
Head of Platform Architecture,
Carestream Health
Ketan Paranjape
General Manager, Life Sciences,
Intel Corporation
Carestream Health’s cloud
leadership and use of Intel®
technologies helps health
systems reduce PACS/RIS
operations costs, manage
data growth, and prepare
for a new era of medical
imaging data analytics.
white paper
Preparing for Next-Generation
Medical Imaging Data and Analytics
Bioinformatics

Table of Contents
Executive Summary . . . . . . . . . . . . . . 1
Driving the Business Need for
Scale and Cost Containment . . . . . . 1
Enabling Technologies: Cloud
Computing, Big Data Analytics,
and NVMe SSDs. . . . . . . . . . . . . . . . . . 2
Industry-Leading Solutions for
PACS/RIS Workflows and
Healthcare Analytics. . . . . . . . . . . . . . 3
Assessing the Impact for PACS. . . . 4
The Next Breakthrough:
3D XPoint™ Technology. . . . . . . . . . . 4
Move Forward. . . . . . . . . . . . . . . . . . . . 5
Reap the Rewards. . . . . . . . . . . . . . . . . 6
Learn More. . . . . . . . . . . . . . . . . . . . . . . 6
At the same time, the volumes and
complexity of medical imaging data
are skyrocketing. By one estimate,
medical images consume up to one-
third of global data storage.2
Study
sizes are expanding. PACS and imaging
databases must handle larger images;
high-definition formats; 3-D video;
and varied modalities such as MRI,
CT, PET, and others. All this data must
be captured, stored, and analyzed by
means of increasingly sophisticated
algorithms. It must be securely shared
in multiple formats and often in near-
real time among authorized members
of the healthcare team, including
patients. It must be tagged intelligently
to enable meaningful analysis, and
it must be archived for long-term
retention. It’s not surprising that medical
imaging storage consumes a rising
percentage of the budget and floor
space in the healthcare data center.
While the expansion of imaging data
creates management and infrastructure
challenges, it also brings significant
opportunities. By integrating medical
imaging data with information from the
electronic health record (EHR), claims
data, scanned documents, genomic data,
lab results, patients’ personal health
records, and other sources, healthcare
organizations establish a rich, dynamic
pool of data from which to gain a
deeper understanding of health and
wellness. When healthcare organizations
succeed in merging Digital Imaging and
Communications in Medicine (DICOM)
and non-DICOM images, bringing
imaging data out of its traditional
silos, and analyzing it effectively, the
resulting analytic insights can help:
• Deliver high-quality, personalized
care and improve outcomes
by presenting providers with a
comprehensive view of patients’
health and history
• Improve patient satisfaction and
outcomes by supporting shared
decision-making and enhancing
patient-engagement strategies
• Increase provider and enterprise
efficiency and reduce costs
• Assess treatment outcomes and
identify innovative diagnostic and
treatment approaches
Enabling Technologies: Cloud
Computing, Big Data Analytics,
and NVMe SSDs
Several major technology advances
can help healthcare enterprises meet
these challenges and take advantage
of the opportunities, primarily cloud
computing. Using cloud services, along
with virtualized local infrastructure,
healthcare enterprises gain a more
scalable, sustainable way to store,
retrieve, archive, and analyze images
and imaging data while reducing
operations costs for PACS/RIS systems.
Well-implemented cloud solutions
also foster secure collaboration among
diverse organizations.
The methods and technologies of big
data analytics provide breakthrough
capabilities for the analysis of medical
imaging data and other unstructured
healthcare data. Big data methodologies
evolved to handle analytics scenarios in
which high volumes of data, of varied
types, are being generated at high
velocities. These scenarios are generally
beyond the capabilities of legacy
infrastructure and relational databases,
so big data technologies use distributed
processing frameworks such as Apache
Hadoop* software, non-relational
analytics databases, new analytics
applications, and distributed
infrastructure. Frost & Sullivan predicts
that data management and analytics tools
for medical imaging in the United States
alone will expand from USD 39 million
in 2014 to USD 82 million annually by
2019, growing at a compound annual
growth rate of 16 percent.3
Solid-state drives, particularly high-
performance drives designed to be
connected through a PCI Express (PCIe)
2Preparing for Next-Generation Medical Imaging
Data and Analytics

bus and accessed using the NVMe
interface, are a valuable infrastructure
element for processing, delivering, and
analyzing images and imaging data.
NVMe is an open industry-standard
interface developed by Intel and other
vendors to take advantage of PCIe’s low
latency and parallelism.
Based on non-volatile flash memory,
enterprise-class SSDs can be attached
to a specific server, server cluster,
workstation, or laptop to accelerate
data movement to and from the
processor. They can also be used in a
shared disk array to alleviate network
bottlenecks. In each case, the SSDs
can significantly increase database and
analytics throughput, performance, and
reliability compared to older SSDs and
traditional SATA hard disk drives, at an
increasingly comparable cost.4
Industry-Leading Solutions
for PACS/RIS Workflows and
Healthcare Analytics
Carestream Health is a leading
provider of medical and dental
imaging and IT solutions, as well
as X-ray imaging systems for non-
destructive testing, and advanced
materials for the precision films and
electronics markets. Its comprehensive
family of web-based and cloud-based
solutions for PACS and RIS, along
with its a “one-workstation”
philosophy of user interface design
and reporting, empower health
organizations to optimize productivity
and maximize investments.
Within the broad Carestream Health
product line, Carestream Vue™ for
Cloud-Based Services is a fully
managed, strategic solution for PACS,
archiving, collaboration, and portal
services, to deliver new ways for
clinicians to collaborate resulting in
better patient care.
Organizations using Carestream Vue for
Cloud-Based Services can:
• Avoid capital investment and reduce
total cost of ownership (TCO) by as
much as 30 percent with predictable,
pay-as-you-go operating costs
• Enjoy the high performance,
flexibility, and reliability provided by
the latest technologies
• Empower radiologists, clinicians,
and referring physicians with access
to imaging information and tools,
anytime from anywhere
• Enjoy a highly available and secure
infrastructure that is monitored and
supported by IT experts with clear
service-level agreements (SLAs)
• Stay in control through proactive
reporting on usage, activity, and
status of data
• Implement a robust, cost-effective
disaster recovery solution
Carestream has a strong focus on
technology innovation and deploys
Intel technologies throughout its IT
infrastructure to provide optimal
performance and reliability for its
customers. It also uses tools such as
Intel® VTune™ Performance Analyzer to
optimize its software environment.
Servers and storage systems powered
by the Intel® Xeon® processor family are
designed for data-intensive workloads,
enterprise-grade clouds, big data
analytics, and data centers of every
size. The newest Intel Xeon processor
E7 v3 family provides energy-efficient
performance, expanded memory
capacity, and enhanced reliability
features. It includes the hardware-
aided virtualization innovations of
Intel® Virtualization Technology
(Intel® VT), offering further support
for software-defined infrastructure.
Intel also drives a broad range of
Intel® SSD Data Center Family
for PCIe* P3700 provided a
dramatic threefold increase
in throughput for a critical
portion of the Carestream
Vue™ workflow. The result:
a more responsive experience
and greater productivity
for clinicians.
Data and Analytics

innovations to enhance data privacy
and security. Intel has also worked with
Cloudera*, provider of the most popular
version of Apache Hadoop software, to
enhance its security, performance, and
management capabilities.
Converged networks with Intel®
Ethernet Controllers offer industry-
leading I/O virtualization innovations
and performance. These controllers
extend Intel VT to the network, to
further enhance the software-defined
data center and allow rapid
provisioning of networks.
The Intel Solid-State Drive Data Center
Family for PCIe brings extreme data
throughput directly to Intel Xeon
processors with up to six times faster
data transfer speed than 6 Gbps SAS/
SATA SSDs.5
The performance of a single
drive from the Intel SSD Data Center
Family for PCIe, specifically the Intel
SSD Data Center P3700 Series (460K
IOPS), can replace the performance of
seven SATA SSDs aggregated through a
host bus adapter (approximately 500K
IOPS). Intel SSD Data Center Family
products also consume less energy than
traditional drives, further reducing the
data center’s environmental impact.
Assessing the Impact for PACS
Technologists from Carestream
Health and Intel established a proof-
of-concept system to explore the
benefits of the new Intel SSDs. The
team incorporated the Intel Data Center
P3700 Series featuring NVMe into the
workflow for database management
and PACS image caching.
The results were dramatic. For a data
set with 4,000–5,000 CT exams,
Carestream Health found that the
configuration shown in Figure 1
provided a threefold increase in
performance, loading 700 images per
second compared to an average of
150 to 200 images per second on the
SAS and SATA hard drives with 1 Gbps
network connectivity.
These increases in performance
can make the difference between a
radiology study that loads in minutes
and one that loads in seconds.
This performance can enhance the
productivity of health professionals,
reducing the time needed to read a
study. It can also help keep pace even
as studies grow more data-rich, with
larger, higher-resolution images and
more images per study. It is an essential
enabler to meeting current and future
PACS/RIS workloads.
The Next Breakthrough:
3D XPoint™ Technology
Further innovation is on the horizon.
Carestream Health continues to
advance its Vue family, and Intel is
pushing its platform technologies
forward in performance, throughput,
capacity, security, and other areas. In
addition, Intel and Micron Technologies
have announced a revolutionary
storage/memory innovation that will
be highly relevant for medical imaging
data and analytics, and other data-
intensive areas. Intel® 3D XPoint™
Technology (pronounced 3D cross-
point) is a disruptive innovation in
non-volatile memory technology that
Data and Analytics
Stress Tester
Client
12 vCPUs
48 GB of Ram
VMXNET3 NIC
Microsoft Windows Server
2012 R2*
Intel® Xeon® Processor E5-2699 v 3 2.30 GHz
96 GB of RAM
Intel® Ethernet Controller XL710 (40 Gbps)
Microsoft Windows Server 2012 R2*
Intel® Core™ i5 Processor
12 GB of RAM
Realtek* PCIe GBE Family
Controller NIC
Windows* 7
Intel® SSD DC P3700 Series for NVMe
NVMe SSDPEDMD016T4 (Database)
Intel® SSD DC P3700 Series for NVMe
NVMe SSDPEDMD016T4 (Images)
PACS
Figure 1. PACS Infrastructure with Intel® SSD Data Center Family for PCIe P3700

is 8–10 times denser than conventional
DRAM memory and up to 1,000
times faster than NAND, the most
popular non-volatile memory in the
marketplace today.6
Intel, Carestream
Health, and other industry leaders are
also continuing work to deliver practical
advances in areas such as imaging
informatics algorithms, collaboration
capabilities, and user interface design.
These innovations will enable
healthcare organizations to support
more users with higher performance
and throughput while controlling costs,
managing data growth, and exploiting
analytics opportunities.
Move Forward
With the maturing of cloud computing,
rapid innovations in big data analytics
methods, a new generation of SSDs
based on NVMe, and the promise of
3D XPoint Technology, it’s time for
healthcare organizations to prepare for
the new era of cloud-enabled medical
imaging data and big data analytics,
if they’re not doing so already. We
recommend a three-part approach.
Develop a Cloud Roadmap
Scalability, performance, throughput,
and security will be top priorities for
medical imaging data in the cloud,
whether on-premises or external. To
build success, select a cloud-based
solution such as Carestream Vue that
provides comprehensive support for
PACS/RIS workflows, and is compatible
with or can interface smoothly with
your web-based imaging applications.
Establish policies that identify which
workloads and data sets will migrate
into a cloud environment. If you decide
to use external cloud services, choose a
partner such as Carestream Health that
has specialized expertise in designing
infrastructure and processes for
securing imaging/healthcare data in
the cloud. A vendor such as Carestream
that is aggressive in adopting the latest
Intel technologies demonstrates its
commitment to maintaining industry-
leading performance and helping
you keep pace with the changing
requirements for medical imaging
data. An experienced healthcare cloud
services partner can also provide
guidance as you determine which
applications and which types of data
need to remain on-premises.
Modernize Onsite Infrastructure
Optimized infrastructure based on Intel
technologies is the foundation for a
balanced, software-defined data center
that provides scalable, secure, and
reliable performance for on-premises
clouds and applications. Having a
modern local infrastructure also allows
for a much more robust interface with
external cloud services.
Modernizing older infrastructure with
Intel technology-based solutions can
improve energy efficiency and density
in the data center while reducing
complexity and enhancing IT efficiency.
Intel technology-based servers, storage
systems, network adapters, and data
center-class SSDs are designed for large
data volumes, demanding data analytics,
and mission-critical enterprise
computing environments. In particular,
incorporating Intel Data Center SSDs
with NVMe can significantly increase
throughput and responsiveness for
PACS/RIS applications, performance-
constrained databases, and other
data-intensive applications. Along
with high-performance servers and
storage systems, your medical imaging
infrastructure will benefit from 40 Gb
Intel Ethernet Adapters to connect
the server backbone, and 10 Gb Intel
Ethernet Adapters for client workstations.
Create and Implement an
Analytics Strategy
An analytics strategy should address
medical imaging data and other types
of healthcare data. It should result
from collaboration among clinical,
By integrating medical imaging
data with EHR information,
claims data, scanned
documents, genomic data,
patients’ personal health
records, and other sources,
healthcare organizations
establish a rich, dynamic pool
of data from which to gain
a deeper understanding of
health and wellness.
Data and Analytics

operational, business, and IT leaders
to identify opportunities where
actionable, data-fueled insights can
improve operational excellence, clinical
care, and/or the bottom line.
With a strategic direction in place, you
can start planning the necessary steps
to improve data interoperability, acquire
or develop new skills, and remove
cultural and technical barriers to data
sharing/integration. Identifying practical
use cases and undertaking small-scale
proof-of-concept projects let you start
gaining value as you build expertise.
Distributed infrastructure for big data
analytics requires the same high levels
of energy-efficient performance,
throughput, and scalability as other
infrastructure for data-intensive
healthcare computing. On the software
side, choose tools and solutions that
maintain open source compatibility
throughout the Apache Hadoop
software stack while providing a secure,
manageable analytics environment.
Reap the Rewards
Delivering high-quality, value-based
care requires cost-effective operations
and data-driven, collaborative
decision-making. It requires managing
medical imaging data efficiently,
sharing it effectively, and increasing the
productivity of health professionals
who create, use, and share it. It requires
analyzing imaging and other healthcare
data, and deriving maximum value from
that analysis. These are not easy tasks,
but they are vital ones.
Intel and Carestream Health are driving
innovations that help healthcare
organizations make the most of their
medical imaging data and achieve
their objectives. By taking advantage
of these innovations—including
Intel’s next-generation platform and
storage technologies, and Carestream
Health’s rapid adoption of Intel’s
innovations—health systems can better
manage PACS/RIS data growth while
optimizing their PACS/RIS and analytics
performance. They can meet today’s
workflow needs, move forward with
medical imaging data analytics, and
free resources to focus on improving
the quality and cost of care.
Learn More
Learn more about how Intel and
Carestream can help you make the
most of medical imaging data. Talk
to your Intel or Carestream Health
representatives, or visit us on the Web:
• www.intel.com/healthcare/bigdata
• www.intel.com/healthcare/
optimizecode
• http://www.carestream.com/
medical.html
Stay abreast of continued innovations.
Follow us on Twitter:
• @portlandketan, @IntelHealth
• @carestream
Join the Intel Health Life Sciences
Community for ongoing updates,
discussions, and more.
Preparing for Next-Generation Medical Imaging
Data and Analytics
1
This performance was derived from a system using Intel SSD DC P3700 Series compared with a system using SAS 15k drives configured in RAID 10 for the database plus either SAS 10k drives or SATA 7.2k drives
configured in RAID 5/6 for image caches.
2
Digital Imaging in Medicine: Trends and Challenges, International Hospital Equipment and Solutions, http://www.ihe-online.com/feature-articles/digital-imaging-in-medicine-trends-and-challenges/index.html .
3
Frost Sullivan, US Medical Imaging Industry Leaps Firmly into Big Data Realm, June 29, 2015. http://ww2.frost.com/news/press-releases/us-medical-imaging-industry-leaps-firmly-big-data-realm/
4,5
Tests document performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. Configurations: Performance claims
obtained from data sheet, sequential read/write at 128k block size for NVMe and SATA, 64k for SAS. Intel SSD DC P3700 Series 2 TB, SAS Ultrastar® SSD1600 MM, Intel SSD DC S3700 Series SATA 6 Gbps. Intel
Core i7-3770K CPU @ 3.50 GHz, 8 GB of system memory, Windows Server* 2012, IOMeter. Random performance is collected with 4 workers each with 32 QD.
6
Technology claims are based on comparisons of latency, density, and write-cycling metrics among memory technologies recorded on published specifications of in-market memory products against internal
Intel specifications. Learn more about 3D XPoint technology at http://www.intel.com/content/www/us/en/architecture-and-technology/3d-xpoint-technology-animation.html and http://newsroom.intel.com/
community/intel_newsroom/blog/2015/07/28/intel-and-micron-produce-breakthrough-memory-technology
Intel® Virtualization Technology requires a computer system with an enabled Intel® processor, BIOS, and virtual machine monitor (VMM). Functionality, performance, or other benefits will vary depending on
hardware and software configurations. Software applications may not be compatible with all operating systems. Consult your PC manufacturer. For more information, visit http://www.intel.com/go/virtualization
Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. Check with
your system manufacturer or retailer or learn more at intel.com
This document and the information given are for the convenience of Intel’s customer base and are provided “AS IS” WITH NO WARRANTIES WHATSOEVER, EXPRESS OR IMPLIED, INCLUDING ANY IMPLIED
WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NONINFRINGEMENT OF INTELLECTUAL PROPERTY RIGHTS. Receipt or possession of this document does not grant any
license to any of the intellectual property described, displayed, or contained herein. Intel® products are not intended for use in medical, lifesaving, life-sustaining, critical control, or safety systems, or in nuclear
facility applications.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific
computer systems, components, software, operations, and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you
in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information, go to http://www.intel.com/performance
© 2016, Intel Corporation. All rights reserved. Intel, the Intel logo, Intel Core, 3D XPoint, VTune, and Xeon are trademarks of Intel Corporation in the U.S. and other countries.

Intel next-generation-medical-imaging-data-and-analytics

More Related Content

What's hot (20)

Viewers also liked (10)

Similar to Intel next-generation-medical-imaging-data-and-analytics (20)

More from Carestream (20)

Recently uploaded (20)

Intel next-generation-medical-imaging-data-and-analytics