Skip navigation
1 2 3 Previous Next

Intel Health & Life Sciences

369 posts

 

The shift from fee-for-service to fee-for-performance is changing the conversation around patient care. Reducing readmissions is one benchmark for analyzing the quality of care, and more discussion is happening around bringing telehealth into the mix to improve this metric.

 

Traditionally, when patients leave the clinical setting, interaction between the care team and the patient decreases. With telehealth and remote patient monitoring, technology allows the provider team to remain in contact with the patient to follow up on regiments and make sure instructions are followed. The result can be a shift in outcomes for the better.

 

To learn more about telehealth, we sat down with Fadesola Adetosoye from Dell Healthcare Services, who says telehealth allows patients to overcome challenges, like transportation issues, to obtain better primary care and stay in touch with clinicians following discharge.

 

Watch the video above and let us know what questions you have about telehealth? Is your organization using a telehealth strategy?

Mark Caulfield, FMedSci, is a chief scientist and board member at Genomics England, an organization which provides investment and leadership to increase genomic testing research and awareness. Caulfield is also the director of the William Harvey Research Institute and was elected to the Academy of Medical Sciences in 2008. His particular areas of research are Cardiovascular Genomics and Translational Cardiovascular Research and Pharmacology. We recently sat down with him to discuss genomic sequencing as well as insight into a current research project. Mark-Caulfield-11-use.jpg

 

Intel: What is the most exciting project you’re working on right now?

 

Caulfield: The 100,000 Genomes Project is a healthcare transformation program that reads through the entire DNA code using whole genome sequencing. That’s 3.3 billion letters that make you the individual you are. It gives insight into what talents you have as well as what makes you susceptible to disease. My research is focused on infectious disease and rare inherited diseases such as cancer. Technology can bring answers that are usable in the health system now across our 13 centers.

 

When studying rare disease, the optimal unit is a mother, father and an affected offspring. The reason is that both parents allow the researcher to filter out rare variations that occur in the genetic code that are unrelated to the disease, focusing in on a precise group. This project will result in more specific diagnosis for patients, a better understanding of disease, biological insights which may pave the way for new therapies and a better understanding of the journey of patients with cancer, rare disease and infection.

 

Intel: How does this project benefit patients?

 

Caulfield: By building a picture of the entirety of the genome or as much as we can read today, which is about 97.6 percent of your genome, we have a more comprehensive picture and a far greater chance of deriving healthcare benefits for patients. Cancer is essentially a disease of disordered genome. With genomic sequencing, we can gain insights into what drove the tumor to occur in the first place, what drives its relapse, what drives its spread and other outcomes. Most importantly, we can understand what drives response to therapy. We already have good examples of where cancer genotyping is making a real difference to therapy for patients.

 

Intel: What is the biggest hurdle?

 

Caulfield: Informed consent is essential to the future application of the 100,000 genomes project. It’s very hard to guarantee, that you can absolutely secure data. I think it’s the responsibility of all medical professionals like myself in this age to be upfront about the risk to data access. Most patients understand these risks. We try and keep patient data as secure as is reasonably possible within the present technological bounds.

 

Intel: What is crucial to the success of genomic sequencing?

 

Caulfield: We need big data partners and people who know how to analyze a large amount of data. We also need commercial partners that will allow us to get new medicines to patients as quickly as possible. That partnership, if articulated properly, is well received by people. Once we have this established, we can make strides in gaining and keeping public and patient trust, which is crucial to the success of genomic sequencing.

 

If you want public trust, you must fully inform patients about the plan. Ensure their medical professionals understand that plan and that patients are bought into a conversation. This allows the patients and the public to shape your work. Sometimes in medicine, we become a little remote from what the patient wants when in actuality, this is their money. It should be their program, not mine.

 

Intel: What goal should researchers focus on?

 

Caulfield: With this large amount of data comes the need to process it as quickly as possible in order to provide helpful results for both the patient and care team. Intel’s All in One Day initiative is an important goal because it accelerates the time from when a person actually enrolls in such a program to receiving a diagnostic answer.

 

The goal is to get the turn-a-round as fast as possible. For example, if a patient has cancer, that person may have an operation where the cancer is removed. Then the patient would then need to heal. If chemotherapy were needed, it would be important to start that as quickly as possible. We have to use the best technology we have available so we can shrink the time from involvement to answer.

 

All In One Day by 2020 – the phrase encompasses our real ambition here at Intel to empower researchers to give clinicians the information they need to deliver a targeted treatment plan for patients in just one 24-hour period. I wanted to provide you with some insight into where we are today and what’s driving forward the journey to All In One Day by 2020.

 

Genomics Code Optimization

 

We have been working with industry-leader experts, and commercial and open source authors of key genomic codes for several years on code optimization to ensure that genome processing runs as fast as possible on Intel®-based systems and clusters. The result is a significant improvement on the speed of key genomic programs which will help get sequencing and processing down to minutes, for example:

 

  • Intel has sped up a key piece of the Haplotype Caller in GATK, the pairHMM kernel to be 970x faster for an overall 1.8x increase in the pipeline performance;
  • The acceleration of file compression for genomics files, e.g. BAM and SAM files by over 4x
  • The acceleration of Python using Intel's Math Kernel Library (MKL) producing a 15x speedup on a 16-core Haswell CPU;
  • Finally, using the enhanced MKL, in conjunction with its Data Analytics Acceleration Library (DAAL), has enabled DAAL to be 100x faster than R for k-means clusters and 35x faster than Weka on Apriori.

 

You can find out more about Intel’s work in code optimization at our dedicated Optimized Genomics Code webpage.

 

Scalability for Success

 

As we see an explosion in the volume of available data the importance of being able to scale a high performance computing system becomes ever more critical to accelerating success. We have put forth the Intel® Scalable System Framework to guide the market on the optimal construction of an HPC solution that is multi-purpose, expandable and scalable.

 

Combining the Scalable System Framework with optimized life sciences codes has resulted in a new, more flexible, scalable, and performant architecture. This reduces the need for purpose-built systems and instead offers an architecture that can span a variety of diverse workloads while offering increased performance.

 

Another key element of an architecture is the balance between three key factors: compute, storage, and fabric. And today we see the fruits of our work coming to life, for example, in a brilliant collaboration between TGen, Dell and Intel which optimized TGen’s  RNA-Seq pipeline from 7 days to under 4 hours. TGen are successfully operating FDA-approved clinical trials, balancing research and providing clinical treatment of pediatric oncology patients.

 

The intersection of our code optimization efforts and our SSF effort have yielded two new products for genomics too, one from Dell and another from Qiagen.

 

From a week to a day

 

It’s useful, I think, to see just how far we’ve come in the last four years as we look ahead to the next four years to 2020. In 2012 it took a week to perform the informatics on a whole human in a cloud environment going from the raw sequence data to an annotated result. Today, the time for the informatics had decreased to just 1 day for whole genomes.

 

With the Dell and Qiagen reference architectures that are based on optimized code and the Intel® Scalable System Framework, a throughput-based solution has been created. This means that when fully loaded these base systems will perform the informatics on ~50 whole genomes per day.

 

However, it is important to note the genomes processed on these systems still take ~24 hours to run, but they are being processed in a highly parallel manner. If you use a staggered start time of ~30 minutes between samples, this results in a completed genome being produced approximately every 30 minutes. For the sequencing instrumentation, Illumina can process a 30x whole human genome in 27 hours using its “rapid-run mode”.

 

So, in 2016, we can sequence a whole genome and do the informatics processing in just over 2 days (51 hours consisting of 27 hours of sequencing + 24 hours of informatics time), that’s just ~1 day longer than our ambition of All In One Day by 2020.

 

Three final points to keep in mind:

 

  1. There are steps in the All In One Day process that are our outside of the sequencing and the informatics, such as the doctor's visit, the sample preparation for sequencing, the genome interpretation and the dissemination of results to the patient. These steps will add additional time to the above 51 hours.
  2. The reference architectures are highly scalable meaning a larger system can do more genomes per day. 4 times the nodes produce 4 times throughput.
  3. There are enhancements still to be made. For example, streaming the output from the sequencer to the informatics cluster such that the informatics can be started before the sequencing is finished will further compress the total time towards our all-in-one-day goal.

 

I’m confident our ambitions will be realized.

 

Telemedicine is gaining increased attention worldwide as a solution for improving access to care, improving quality of care, and lowering costs.

 

Much of Latin America faces a major challenge that could in part be addressed with telemedicine:  a shortage of providers, and large populations living in rural areas where access to physicians—particularly specialists—is lacking.

 

In my multiple visits to Latin America over the past two years, it is clear that while most countries in the region have used telemedicine to varying extents for many years, scalability remains a major goal.

 

Governments across Latin America are generally strong advocates of telemedicine, and are investing in the networks and infrastructure that will support this technology.

 

Below I highlight ways in which countries throughout the region are using or intend to use telemedicine, and what trends we might observe in the years ahead.

 

Brazil

In Brazil, telemedicine today is used strictly for provider-to-provider consultation, as physicians are not legally allowed to consult with patients over videoconference.

 

Telemedicine has been largely driven by the need to provide care virtually between specialists in urban centers to patients in remote areas, due to a lack of specialists in the rural areas.

 

The Brazilian government has long supported the use of telemedicine to provide better access and treatment to remote areas. Since 2006, it has facilitated two public initiatives--the Brazilian National Telehealth Network Program (launched by the MOH) and the RUTE-Telemedicine University Network (launched by the Ministry of Science, Technology, and Innovation) both of which serve to deploy telemedicine across Brazil.

 

One of the first major initiatives started in 2006 in Parintins, a city of 100,000 located in the middle of the Amazon. With no roads to or from the city, the goal was to use telemedicine to enable communication between physicians in Parintins and specialists in Sao Paulo. Parintins partnered with private technology companies, including Intel, to build the necessary infrastructure (e.g., WiMAX network). This telemedicine program continues to operate today, and has informed other telemedicine efforts including Brazil’s national telehealth program, Telessaude (http://www.telessaudebrasil.org.br/).

 

Another major initiative in Brazil is to bring intensive care unit (ICU) care to rural areas. The Brazilian MOH initiated tele-ICU programs so that now many hospitals in different regions are connected to rural parts of the country. These tele-ICUs reduce the need to transport patients into a city for health conditions such as heart attacks, strokes, and sepsis. Physicians in urban areas are able to use PTZ cameras to visually inspect the patient, and collect and interpret vital signs in real-time. Cerner, in partnership with Brazilian companies Intensicare and IMFtec, has provided the technology and software for most of these virtual ICUs.

 

Mexico, Chile, Peru, and Argentina

In Mexico, the social security network provides healthcare to formal sector workers. The network is currently working with companies such as Lumed Health http://www.lumedhealth.com/ to expand telemedicine capabilities. In addition, telemedicine is being used between the U.S. and Mexico with health systems such as the Mayo Clinic and Massachusetts General conducting consultations with physicians in Mexico.

 

In Chile, the Ministry of Health has implemented a “Digital Health Strategy.” Its primary goal is also to address provider shortages and to improve access to care in rural areas. There are currently several telemedicine projects and POCs underway in Chile.  AccuHealth (https://www.accuhealth.cl/), for example, is a Chilean company that provides tele-monitoring services specifically to bring home care to patients who suffer from chronic conditions. The company plans to expand to Mexico and Colombia in the near term.

 

In Peru, the government is spearheading efforts to build a fiber optics network across the entire country (www.proinversion.gob.pe/RedDorsal/). This infrastructure will be used to better support telemedicine services.

In Argentina, the government has worked with the MOH and the Ministry of Federal Planning, Public Investment and Services to promote telemedicine. This collaboration has culminated in the CyberHealth Project, which is focusing on the installation of fiber optics and upgrading hospitals to allow for videoconferencing. It aims to connect 325 healthcare institutions across the nation to enable remote consultations and sharing of expertise.

 

The Future of Telemedicine in Latin America

Telemedicine is being increasingly recognized as a solution to achieve more with less. In Latin America, it has great potential to address the fact that providers and health care resources are not distributed equally among the urban and rural populations.

 

The future of telemedicine in the region is promising. Governments are investing in and taking active roles in digitizing their health systems (e.g., implementation of electronic medical records, improving interoperability) along with building the infrastructure required to support telemedicine. The Pan American Health Organization (PAHO) has convened a meeting of the MOH leaders from several Latin American countries to discuss strategic plans for e-Health across the region. This collaboration, where protocols, guidelines, and best practices can be shared, will be increasingly important.

 

Intel Health & Life Sciences looks forward to continuing its partnerships with public and private entities across Latin America to continue these important efforts.

International Nurses Day is a time to say Thank You Nurses. Thank you for your hard work, thank you for your compassion and thank you for the endless care you give to patients. It’s this unwavering focus on patient care that we must keep in mind when developing and implementing technology for nurses both in the hospital and community. The most valuable technology we can give to nurses is that which is almost invisible to - yet improves - their workflow, simplifies complex tasks and enables them to deliver even better care – in essence, technology must make the job of a nurse easier. I want to take today, International Nurses Day, to highlight a couple of technologies which have the potential to deliver on all of the above.

 

Nursing goes Digital

I know from experience that the best decisions are made when a nurse has the most accurate and up-to-date information on a patient’s condition. And when that accurate information can be gathered and accessed in an intuitive and more natural interaction using technology it’s a win-win for nurses and patients.

 

I’m excited by the potential offered by Intel’s RealSense 3D camera which can be found in a range of devices such as 2-in-1s, the likes of which are already being used by nurses to record vital signs and access EMRs. For example, imagine being able to accurately track all 22 joints of a hand to assist with post-operative treatment following hand surgery.

 

For community nurses, mobility is key. Holding the most up-to-date information when visiting patients in the home ensures mistakes are kept to a minimum and all parties involved in the care of the patient, from community nurses to specialist clinician, can make evidence-based decisions. 2-in-1 devices help nurses to stay focused on the patient rather than reams of paperwork, while also helping patients better understand their condition and improving buy-in to treatment plans. The real benefits are in simplifying and speeding up those processes which ensures nurses deliver the best possible care.

 

Big Data for Nurses

When we think of Big Data it is all too easy to think just about genomics, but there are benefits which can clearly help nurses identify serious illness more quickly too. Take Cerner for example, who have developed an algorithm that monitors vital information fed in real-time from the EMR. The data is analysed on a real-time basis, which then identifies with a high degree of accuracy that a patient is either going to get, or already has, sepsis.

 

Clearly, given the speedy nature with which drugs must be administered, this Big Data solution is helping nurses to simply save lives by identifying at-risk patients and getting them the treatment they so desperately need. Watch this video to find out more about how Intel and Cloudera allow Cerner to provide a technology platform which has helped save more than 2,700 lives.

 

Intelligent Care

The rise of the Internet of Things in the healthcare sector is seeing an increasing use of sensors to help simplify tasks for nurses. For example, if sensors can monitor not only a patient’s vital signs but also track movement such as frequency of the use of a toilet, it not only frees up a nurse’s time for other tasks but also begins to build an archive of data which can be used at both patient and population effort.

 

In China the Intel Edison-based uSleepCare intelligent bed is able to record a patient’s vital signs such as rate and depth of breathing, heart-rate and HRV without the need for nurse intervention. There are positive implications for patient safety too, as sensors can track movements and identify when patients might fall out of bed, alerting nurses to the need for attention.

And when I think of moving towards a model of distributed care, this type of intelligent medical device can help the sick and elderly be cared for in the home too. WiFi and, in the future, 5G technologies, combined with sensors can help deliver the right patient information to the right nurse at the right time.

 

Investing in the Future

Having highlighted two examples of how technology can help nurses do an even better job for patients I think it’s important to recognise that we must also support nurses in using new technology. Solutions must be intuitive and seamlessly fit into existing workflows, but I recognise that training is needed. And training on new technologies should happen right from the start of nursing school and be a fundamental part of ongoing professional development.

 

While International Nurses Day is, of course, a time to reflect and say Thank You Nurses, I’m also excited about the future too.

 

Last week I had the honor of moderating the weekly #HITsm (Health IT social media) chat on Twitter. This regular discussion about health IT issues is a wonderful forum for addressing what steps need to be taken to move healthcare technology forward on a number of fronts.

 

The topic of my chat was The Growth of Connected Care, and focused on defining the terms, sharing trends and identifying successful characteristics of a connected care program. I enjoyed the banter and the great questions that came my way during the chat and learned quite a bit about what the climate is like for overcoming obstacles to adopting connected care.  You can see the transcript of the entire chat here.

 

To recap the conversation, below are the questions that were asked during the chat and my brief answers.

andychat.png

 

Connected care is a broad term – what does it mean?

Generally, connected care applies to leveraging technology to connect patients, providers, and caregivers. Increasingly, this is happening in real-time. Connected care extends care outside of the traditional hospital setting and moves healthcare from episodic events to more continuous care that is tailored specifically for the patient.

 

What market trends are driving connected care?

A few trends are driving connected care forward. First, new Internet of Things (IoT) technology (devices-datacenter) are making connected care possible for patients. Think about wearables and the massive amount of data that can be acquired that influences care; this is the cornerstone of connected care.

 

Second, payment reform and payment models are changing from fee-for-service to value-based. As payment models change, patient retention becomes increasingly important for clinicians. This is the consumerization of healthcare, where the patient takes charge of their own health and the care is on a regular, on-going basis.

 

Finally, healthcare technology investments in digital platforms have opened the opportunity to create and consume new data streams in real-time.

 

What technologies are enabling connected care?

For starters, big data technologies, both software and hardware, are enabling us to work with the high volume, variety, and velocity of connected care data. Wearables and sensors are also evolving, and newer devices are delivering more value in improved form factors.

 

What are characteristics of a successful connected care program?

Successful connected care programs have clear clinical and business goals, know the problems that need to be solved, have measurable outcomes and clear value propositions, and feature scalable architecture for data ingestion, storage, analysis, and visualization.

 

Programs must be patient-centric and look holistically at both patient and care team touch points throughout the continuum of care. They also need a strategy for transforming data into actionable/comprehensible insights delivered at the right time, to the right person. This is often overlooked – insights for providers or patient instructions get lost in poor visualization. This is why the UI/UX aspect of connected care is so critical.

 

Where is connected care headed, and what are some things to watch for?

Expect larger connected care programs with employers, payers, and care providers to reach consumers and tie engagement to financial outcomes. It will be interesting to see how employees respond and how the employer/employee relationship is re-written to include health-related activities.

 

Population health programs will go through a three step evolution of understanding, predicting, and then preventing (UPP). Step one is simply understanding what data is available and identifying/filling gaps. The second stage of program maturity involves using data to being predicting outcomes for specific populations. This stage involves iterating through models to improve specificity both for target outcomes and population boundaries.

 

The third stage is using the predictions to implement real programs that prevent target outcomes from occurring. This stage will partially rely on human-centered care delivery, but it will also push the boundaries of virtual medicine in response to access and delivery constraints that inevitably arise.

 

On the downside, large data breaches look inevitable in the future as more devices allow for more attack vectors. The big unknown is how this will impact the industry and consumers.

 

What are some of the short- and long-term obstacles to adoption of connected care programs?

The business models for connected care are still evolving. New payment and reimbursement pathways are needed to create growth. Sustainable, long-term patient engagement is a challenge. Hopefully, healthcare will continue to look to industries that have pioneered techniques for data-driven high-touch consumer engagement (consumer goods, SaaS internet companies, etc.) and apply those learnings to developing new strategies to engage patients. Finally, federal and state regulation must continue to evolve because connected care operates across traditional geographic boundaries and models of care delivery.

Nurses Week is a great opportunity to celebrate all of the fantastic work we do for patients. I often find myself pausing at this time of the year to appreciate just how different – and in most cases better – our working practices, processes and outcomes are compared to just 10 or so years ago. Technology has been a great enabler in improving the workflow of nurses today, but I wanted to share some thoughts on the future of nursing in this blog and how we might be welcoming avatars and the world of virtual reality to Nurses Week celebrations in the near future.

 

Better Training, Overcoming Global Shortage of Nurses

There are challenges ahead for the nursing community, driven by many of the same factors affecting the entire healthcare ecosystem, ranging from an increasingly ageing population to pressure on budgets. When I met with nurses from across Europe in Brussels earlier this year at Microsoft in Health’s Empowering Health event, two key themes really came to the fore:

  • First, there was a call for improved training for nurses to help them better understand and benefit from technologies such as 2 in 1 tablets and advanced Electronic Medical Record systems;
  • Second, there was a discussion around what technologies might help overcome the potential of a global shortage of nurses in the future. A 2015 World Health Organisation report stated that ‘a fundamental mismatch exists between supply and demand in both the global and national health labour markets, and this is likely to increase due to prevalent demographic, epidemiologic and macroeconomic trends.’


Looking ahead I see a real opportunity to integrate avatars and virtual reality into the nursing environment which will not only train students to be better nurses but also deliver better patient care with improved workflows at the bedside

 

Virtual Reality To Deliver Safe, Effective Teaching

Training is a fundamental part of a nurse’s development, and that rings true for both those in nursing school and more experienced nurses learning new technologies and procedures. Virtual reality technology can play a major role in helping nurses to better deal with a range of scenarios and technologies.

 

For example, if I want to teach a nurse how to perform a specific procedure using virtual reality, I’m able to present the trainee with an avatar on a screen that could be any combination of gender, height, weight and medical condition. And whilst the procedure is being undertaken I’m then able to trigger a wide range of responses from the avatar patient to help the nurse learn how to deal with different scenarios – all in a safe and controlled manner that can be monitored and assessed for post-session feedback.

 

Similarly, if a nurse is required to understand how to use a new piece of technology to improve their workflow, such as working with an upgrade to an EMR system on a 2 in 1 tablet, virtual reality can help too by simulating these new systems. In a virtual setting nurses are not only able to familiarise themselves with new processes but can provide feedback on issues around workflow before they are launched into a live patient environment.

 

If I think of how my training was delivered at nursing school there was plenty of ‘chalk and board’-style teaching and a lot of time spent in a classroom using limited resources such as manikins. Today, an infinite number of student nurses can learn remotely using virtual reality and avatar patients, reinforcing knowledge and improving workflows on a range of mobile devices. This is particularly useful too for countries where educators are in short supply but nursing demand may be high.


Avatars Ask ‘How Are You Feeling Today’?

A recurring question in my mind is how can we make better use of the fantastic expertise and knowledge of today’s nurses to continue to deliver great care to patients. In the face of a shortage of nurses we should explore how avatars on a bed-side screen or 2 in 1 device might be able to take away the burden of some of the more routine daily tasks such as asking patients if they have any unusual symptoms.

 

Patient answers could be fed back into the EMR which would trigger either further questions from the avatar or, in more serious cases, an alert for intervention by a human nurse.

When I talk to my peers within healthcare there are some obvious and real concerns about the lack of emotion delivered by avatars. The rise of chat bots has made for interesting news recently and I see this kind of artificial intelligence combined with a great avatar experience delivering something approaching human emotions such as sympathy for routine tasks. We should also recognise that an avatar can potentially speak an unlimited number of languages too, helping all patients get a better understanding of their condition.

 

As nurses I hope our community embraces discussion and ideas around the use of virtual reality and avatars – I’ve talked through just a couple of scenarios where I see improvements to training and care delivery but I’d be interested to hear how you think they could help you do your job better. And perhaps one day in the near future, avatars will be celebrating Nurses Week with us too.

 

abartley

Improve Your Healthcare IQ

Posted by abartley May 4, 2016

Healthcare is undergoing massive changes. As a result of these changes many of those that work in the healthcare industry are finding that they need new skills and knowledge. A great way to go about this is participating in a massive open online course (MOOC).

 

The term MOOC was first used by Dave Cormier of the University of Prince Edward Island in 2008. MOOCs are online courses that are built for open and collaborative participation. MOOC courses are often delivered as a pre-recorded series of video lectures with corresponding assignments to test knowledge. Courses are typically self-paced which makes it easy to schedule around work and family commitments. Mobile applications are available for some platforms which makes learning on the go easy (and much more productive than gaming!). Several MOOC platforms have implemented paid certification programs that focus on in-demand skill sets like data science. In addition to the education, most MOOC platforms provide community forums which can be great ways to connect with other individuals around the world with a shared passion for the subject matter.

 

A variety of healthcare related courses are available on various MOOC platforms. A useful tool for selecting courses across platforms is Mooc List. Three of the more common platforms that come up in healthcare related searches are Coursera, edX, and FutureLearn. Each of these platforms has a slightly different focus in terms of course content and geographic distribution of educators. Coursera seems to have the most diverse set of healthcare curriculum today, but interesting course can be found on all three. Below are some of the sample courses available:

 

Coursera

Interprofessional Healthcare Informatics

We will explore perspectives of clinicians like dentists, physical therapists, nurses, and physicians in all sorts of practice settings worldwide. Emerging technologies, telehealth, gaming, simulations, and eScience are just some of the topics that we will consider.

 

Big Data Analytics for Healthcare

We introduce the characteristics and related analytic challenges on dealing with clinical data from electronic health records. Many of those insights come from medical informatics community and data mining/machine learning community. There are three thrusts in this course: Application, Algorithm and System

 

edX

Entrepreneurship and Healthcare in Emerging Economies

Explore how entrepreneurship and innovation tackle complex health problems in emerging economies.

 

Practical Improvement Science in Health Care: A Roadmap for Getting Results

Course will provide learners with the valuable skills and simple, well-tested tools they need to translate promising innovations or evidence into practice. A group of expert faculty will explore a scientific approach to improvement — a practical, rigorous methodology that includes a theory of change, measurable aims, and iterative, incremental small tests of change to determine if improvement concepts can be implemented effectively in practice.

 

FutureLearn

Inside Cancer: How Genes Influence Cancer Development

In this free online course, you’ll learn about the fundamental biological concepts that inform our current understanding of cancer development, the molecular genetics behind it and its spread within the body.

 

Bioprinting: 3D Printing Body Parts

This free online course tells the story of this revolution, introducing you to commonly used biomaterials, including metals, ceramics and polymers, and how bioprinting techniques, such as selective laser melting, hot-melt extrusion and inkjet printing, work. Through case studies - ranging from hip implants to facial transplants to lab-grown organs - we’ll answer questions such as: What is 3D printing and how did it come about? Is it really possible to print structures that incorporate both living and artificial components? How long before we can print whole body organs for transplants? What is possible right now, and what will be possible in 20 and 50 years’ time?

So whatever your reason, take some time to participate in an MOOC. It’s a fantastic way to stimulate new ideas and connect with like-minded individuals around the world.

 

What questions do you have?

The “Internet of Things” (IoT) has exciting near-term prospects in healthcare.  But what does that mean, and how can we most efficiently realize its potential?

 

Healthcare IoT can take many forms.  Here, we’re referring to sensors deployed onto or inside a human body, that send their data readings to the cloud, which then communicates processed data to clinicians for action.

 

It sounds straightforward, especially if you’re a technologist, because most of the words in the previous sentence are technology words: “sensor,” “data,” “cloud,” “communicate,” and “process.”

 

But notice that other word: “action.”  It’s the last word because it’s the system’s entire reason for being.  If you’re designing your IoT system, and you aren’t clear idea what the actions are, how well they work, and, crucially, how the data are tied to the actions, then pause.

 

What’s Being Tried?

 

Let’s take an example: the recently published BEAT-HF study of heart failure patients.  All patients got their usual care, but half were randomly selected to additionally get coaching telephone calls plus an IoT solution that acquired daily blood pressure, weight, and oxygen saturation – exactly the parameters cardiologists follow in their heart failure patients.

 

Unfortunately, the trial showed no benefit of the IoT solution.  Compared to the control group, the IoT patients died just as often, and they came into the hospital just as often.  This is not the first trial to show such failures, and it is fortunate that BEAT-HF did not harm the subjects by wasting physician time and distracting them from interventions that could actually benefit patients.

 

A Better Mouse-Trap

 

But now let’s look at a different system, also aimed at heart failure patients.  Here, a small Bluetooth-enabled pressure sensor is placed into the pulmonary artery via catheter.  (Pressure in the pulmonary artery is a key indicator of heart failure.)  Once a day the patients lies quietly in bed, near a Bluetooth receiver, and the sensor’s measurements of pulmonary artery pressure are sent to the cloud, and then to the cardiologist’s office.

 

In a randomized study of 550 patients, the patients who received the pressure sensor had their medications changed by the cardiologist 250% more times than the control group.  That is not a typo – 250% -- a remarkable change in the “action” step. But did all that extra “action” help? Yes!  Patients with the pressure system experienced 43% fewer deaths, and 57% fewer heart failure hospital admissions.  The word “spectacular” underestimates this accomplishment, especially given the statistics that, among fee-for-service Medicare enrollees, heart failure is responsible for 39% of all deaths, and for 42% of all hospital admissions.

 

Wrap-up

 

If you are designing an IoT system for healthcare, what lessons can you draw?

 

  • (1) Sensor choice matters.  A lot. Try to obtain data from the core of the disease process, not peripheral or indirect indicators.
  • (2) Merely increasing the data collection frequency, as BEAT-HF tried, may not be beneficial. “Big data” is not a panacea.  Data quantity may not make up for only marginal improvements in data quality.
  • (3) Patient choice matters.  BEAT-HF failed in its general population of heart failure patients, but might have succeeded with certain subgroups of patients.  For example, patients having both heart failure and depression might disproportionally benefit from the Hawthorne effect (increased attention) that telemonitoring can provide.
  • (4) Test your system with a randomized trial.  It is increasingly clear that other study designs are unreliable when evaluating tele-health systems.

 

Although technology terms may dominate the definition of a healthcare IoT system, the single clinical word dominates its success.

Ransomware has reached headlines lately with several healthcare organizations globally falling victim, as seen in As Ransomware Crisis Explodes, Hollywood Hospital Coughs Up $17,000 In Bitcoin. Breaches are top of mind in healthcare as far as security and privacy, and within many types of breaches ransomware is the highest priority across most healthcare organizations I have worked with over the last six months.

 

Compliance with regulations, laws and standards is important, but increasingly organizations realize they need to go well beyond basic regulatory compliance to effectively mitigate risk of breaches, and they are motivated up to the board level with the strong desire to not be the next breach or ransomware victim and headline. Ransomware.jpg

 

While most security concerns to date have revolved around breaches of confidentiality, or unauthorized access to patient information, ransomware is not a breach of confidentiality, but rather of availability. In security speak, “availability” is timely and reliable access to patient information. Ransomware prevents access to patient information by encrypting this information and withholding the decryption key until a ransom is paid. Exacerbating this, paying a ransom is no guarantee of provision of the decryption key.

 

As we have seen, this can compromise mission critical services to where hospitals need to turn patients away. Healthcare is particularly vulnerable to this type of breach because they are generally lagging other verticals in security, and have a very low tolerance for disruption. I suspect this problem is a lot worse than most people realize because many ransomware infections go unreported, as many countries lack breach notification rules, or those rules cover compromise to confidentiality, but not availability as in the case of ransomware.

 

A real danger in securing against this type of breach is the tendency to gravitate to one particular safeguard, such as backup and restore, which while important is just one of many things you can do to secure yourself against ransomware. In this blog, I explore several different safeguards you should consider as part of your holistic, multi-layered, defense-in-depth approach in securing against ransomware. None of these alone is a panacea. Together they represent a very effective, holistic, multi-layered, defense-in-depth security posture against ransomware.

 

  1. Policy: ransomware often starts with employee actions and mistakes. Examples include clicking malicious links in emails or websites, opening email attachments, plugging in malware infected removable storage devices such as USB keys and so forth. Policy governs employee actions. Is your policy accurate, complete and up to date, especially as it pertains to employee actions that can lead to ransomware infections?
  2. Audit and Compliance: policy is a critical foundation of your security practice. To ensure employees are following it you need audit and compliance, in particular to ensure employee compliance with policy in the areas that could lead to ransomware infection.
  3. Risk Assessment: risk assessment is a key tool to identify risks to confidentiality, integrity and availability of patient information, including for risks such as ransomware. You can prioritize risks by impact and probability of occurrence, triage the top risks and address them through application of safeguards. The business impact of ransomware goes well beyond the ransom that may be paid since it can disrupt your mission critical business systems and processes and effectively halt your business.
  4. Anti-malware: having a good anti-malware solution installed on all endpoints, updated and effective is key in detection and remediation, for example quarantine, of malware including ransomware. You will not catch all ransomware this way, but many, especially older variants, will be caught.
  5. User Awareness Training: most ransomware infections start with employee actions. Training can help employees detect and avoid actions that could lead to infections. Again, not a perfect safeguard, but important in your overall anti-ransomware defense. Spear phishing training is particularly important to include in your overall training program.
  6. Email Gateway: email is a key ransomware infection vector, with spear phishing emails containing malicious links coaxing employees to click them, in which case a drive-by-download and infection of ransomware can result. Your email gateway can oversee emails and detect and block many of these.
  7. Web Gateway: web browsing (and clicking) is another key infection vector, with employees visiting websites and inadvertently clicking on malicious links that cause ransomware infections, again by drive-by-downloads. A good web gateway can detect many such websites, and help block these types of infections.
  8. Vulnerability Management and Patching: vulnerable devices and software create openings for malware and ransomware infections. A good vulnerability management program can identify vulnerabilities, for example in old, unpatched, or misconfigured software, and proactively remediate such vulnerabilities to block ransomware.
  9. Security Incident Response Plan: in the event of an infection such as ransomware, how your organization responds is key to faster resolution and minimizing business impact. Having a good, tested plan that employees can execute to quickly and efficiently, with good coordination, is key to enabling this. This plan should include PR and communications for breach notification if needed.
  10. Backup and Restore: currently the “safeguard du jour” for ransomware, backup and restore is critical. Have it, use it (everywhere you have data), test it (test restore regularly), and make sure it is versioned, and some versions air-gapped with offline backup archives. Ransomware may get into your backups too, depending on when it occurs in your backup cycle, and how quickly you detect it and stop it, but if you have versioning and / or an air-gapped backup then you will have a workable backup version to restore. Keep in mind this is not a panacea though, since rolling back to a previous backup version effectively undoes updates since then, and missing patient information updates can translate into direct risks to patient safety and business impact. This is why backup and restore is necessary but not sufficient. It is far preferable to avoid ransomware in the first place.
  11. Device Control: this is the ability to enforce policy regarding removable storage. For example if an employee plugs in a ransomware infected removable storage device such as a USB key, this safeguard can enforce policy preventing ransomware jumping from the device to your IT network.
  12. Penetration Testing and Vulnerability Scanning: as seen in FBI raises alarm over ransomware targeting U.S. businesses ransomware can enter your network through vulnerable or unpatched software, especially software facing the external Internet. Proactive penetration testing such external facing applications and interfaces to identify and remediate such vulnerabilities is key to mitigating risk of this type of ransomware infection.
  13. Endpoint DLP: Data Loss Prevention software running on endpoint devices can enforce policy and help prevent user actions that can lead to malware infections such as ransomware.
  14. Network Segmentation: segmenting your network can help quarantine or localize any malware infections to prevent propagation across your network. This can limit the extent of infection, lessening business impact, and enabling faster resolution.
  15. Network IPS: a network Intrusion Prevention System can monitor network traffic to detect and prevent malicious activity, such as that which could lead to a ransomware infection.
  16. Whitelisting: useful on endpoint devices, whitelisting limits which applications can execute to a small list of approved applications. If ransomware was to get onto a machine with whitelisting it would be benign on that machine since it is not on the approved list of applications and therefore blocked from executing, and therefore unable to encrypt any patient information. This type of safeguard can be particularly useful on medical devices that don’t get patched or updated frequently.
  17. Network DLP: this type of DLP runs on a network and can enforce policy, including detection and prevention of network interactions and traffic that could lead to ransomware infection.
  18. Digital Forensics: in the event of an infection, digital forensics can help identify the type of ransomware, the extent of infection, and how it occurred, which are key to reducing business impact, and preventing future infections.
  19. SIEM: Security Information and Event Management can help provide realtime analysis of security alerts from across your applications and network, enabling faster detection and remediation of ransomware.
  20. Threat Intelligence Exchange: this can enable realtime exchange of threat information between safeguards in your network, and a global threat intelligence backbone from your security provider(s), helping orchestrate defense against ransomware. This is a critical part of the “immune response” of your organization to ransomware, which will help stop it and kill it as fast as possible.
  21. Business Continuity and Disaster Recovery: as we have seen some recent high profile ransomware infections have essentially shutdown the information technology systems of healthcare organizations, crippling mission critical business processes to the point where they had to send patients elsewhere. Having a good BC / DR capability with mirroring of data and hot standby can be helpful in keeping mission critical systems going while remediation is occurring. The effectiveness of this safeguard against ransomware depends on ransomware not propagating to your hot standby system, as can be prevented by various safeguards discussed previously.

 

No organization wants to be “at the back of the herd” or “low hanging fruit” for attacks such as ransomware. It has been difficult in the past for healthcare organizations to measure or benchmark their breach security against the rest of the healthcare industry. It is one thing having a gap in your safeguards if everyone else has that gap. However, if you have a gap and most others don’t then you could be relatively vulnerable.

 

Intel Health and Life Sciences and several industry partners are currently conducting complementary, confidential breach security assessments for provider, payer, pharma and life sciences organizations globally. Through this one hour engagement healthcare organizations are able to benchmark their breach security across 42 safeguard capabilities and 8 different types of breaches, including ransomware, against the rest of the industry to see what percentile they are in terms of readiness, and gaps and opportunities for improvement they may have.

 

Following Bio-IT World, we’re asking some of the world’s top researchers how next generation sequencing (NGS) benefits them and their work. Today, we catch up with Mayo Clinic expert David I Smith, Ph. D, who says NGS allows him to ask and answer questions in a surprisingly short amount of time. The real value, he points out, is that sequencing gives researchers the ability to look at trillions of molecules to see what is happening to populations and move research discoveries, particularly in cancer, forward.

 

Watch the above video to learn more and discover how cancer treatments will be dramatically different five years from now, and what keeps Smith up at night when it comes to NGS.

Realizing the potential in big data is a challenge we’re enthusiastically tackling head on here at Intel and a recently announced strategic partnership with the Alan Turing Institute (ATI) in the UK is just one example of where working with key partners can help us drive scientific and technological discoveries.

 

We want to help turn the rapidly increasing volume of data into meaningful insights which will help solve global challenges across a number of areas, including health and life sciences. The ATI’s vision is an exciting proposition, and that is to be a national institute which supports the UK in becoming a world leader in data science, through:

 

  • Research into the fundamentals of algorithms for data science;
  • Training the next generation of researchers;
  • Addressing ways in which scientific advances can be taken into practice;
  • Collaborating with a range of public and private organizations.

 

If you want the deep dive on the ATI’s forward looking vision, I’d highly recommend reading Andrew Blake’s (Institute Director) Alan Turing Institute Roadmap for Science and Innovation.

 

Alan Turing is a name that is familiar to many of you I’m sure and as the person who many see as the founder of modern computer science we are delighted that new algorithms developed by the ATI will feed into the design of future generations of Intel® microprocessors. Intel will provide the ATI with world-class High Performance Computing solutions including Intel® Xeon®-based workstations, Intel Software tools and access to an Intel Data center cluster based on Intel® Xeon® and Intel® Xeon Phi™.

 

People and Technology

But great technology is just one part of the story of Intel’s strategic partnership with ATI, so I’m excited to tell you that we’re supporting the development of the next generation of data scientists too. Alongside hiring a number of talented individuals to work at the ATI we will be supporting the PhD and Research Fellow programme which will help fulfil one of the core aims of the Institute in helping to bridge the skills gap and place the UK in a strong global position in this sector.

 

Solving the Big Data Challenges in Healthcare

Analysis of big data has the potential to solve some of the biggest challenges in healthcare which will help us deliver better patient care, including All-in-One-Day personalized medicine, unlocking the value of electronic medical records through natural language processing and making sense of the ever-increasing data produced by wearables and sensors. It’s an exciting time and we’re eager to see where this fantastic strategic partnership between Intel and the Alan Turing Institute takes us in the coming years. I look forward to keeping you updated in future blogs.

 

Intel is driving toward a day when cancer patients routinely have their tumor DNA sequenced and receive precision treatment plans based on their unique biomolecular profile—all within 24 hours. We call this vision All in One Day, and we believe that with the right blend of industry-wide commitment, innovation, and collaboration, we can deliver on that vision in 2020.

 

All in One Day isn’t an endpoint, though. I believe it’s also a step toward a world in which life science researchers use ultra-sophisticated 3-D models to simulate the workings of the human body and predict health outcomes. As Dr. Jason Paragas, director of innovation at Lawrence Livermore National Lab, likes to say, “We’d never ask an engineer to build a bridge or design an airplane without modeling how it’s going to perform in the real world. But doctors do the equivalent every day.”

 

If we can empower researchers with advanced biomedical models and simulations, we stand to transform the practice of medicine. Building on the genomics revolution, we may be able to take much more guesswork out of medicine and dramatically expand the universe of available diagnostics, treatments, and preventive approaches.

 

It’s going to take massive increases in computing performance to support these breakthroughs. In the United States, the President’s National Strategic Computing Initiative (NSCI) aims to advance the technologies needed for computers that are 100 times more powerful than today’s most capable supercomputers. Other nations are moving forward with similar initiatives.

 

I recently worked with two of my HPC colleagues to develop a whitepaper that explores precision medicine and discusses Intel’s role in enabling it.

 

We talk about the central role of Intel® Scalable System Framework and its ability to support the convergence of HPC modeling/simulation, health analytics, machine learning, and visualization that precision medicine will require.

 

We touch on key technology innovations as well as collaborations with life science leaders to create open source platforms, tools, applications, and algorithms for precision medicine.

 

And we note that the advances provided by these extreme-scale computers will help us address critical challenges like climate change and renewable energy sources as well as enabling progress toward predictive biology and precision medicine.

 

I hope you’ll read the whitepaper and share your thoughts. What opportunities do you see for life sciences computing to transform biomedicine?  What roadblocks are in the way?

                                                            

Ransomware, it’s a word I’m seeing with increasing frequency amongst security experts. And it’s one I’m keen to let others know about within healthcare because the dangers are already having a major impact on organisations in health and life sciences. A couple of months ago it was reported that a hospital in Germany suffered a security breach which led to all Electronic Medical Records being locked in what at first appeared to be a ransomware attack, with the hospital confirming that the malicious virus had been sent from an unknown source. Fortunately, in this case, the hospital added that no patient information had been accessed but they had not yet calculated the cost to the organisation in regaining access to the data.

 

Ransomware Could Cripple The Ability to Deliver Care

When you consider that a personal health record can be 10x to 20x more valuable to a criminal than an individual’s credit card information you begin to understand the scale and importance of mitigating a wide range of security breaches for healthcare originations. Breach types like ransomware compound unauthorized access to sensitive patient information, compromising the ability of healthcare providers to access this information and crippling their ability to deliver care. No organisation is immune from breaches.

 

Security Workshop for Nordic Regions

That’s why I’m excited to welcome security experts from Intel, including David Houlding, Intel’s Healthcare Privacy and Security Lead, to Sweden at the end of May 2016 for a workshop to help healthcare organisations gain a better understanding of their breach security maturity, and benchmark their priorities across 8 breach types including ransomware, as well as 42 breach security capabilities with the rest of the health and life sciences industry. The event is invite only but if you are interested in finding out more on behalf of your healthcare organization and potentially attending please do get in touch today.

 

At the workshop, David will be talking through and helping organisations get the most out of the Security Maturity Model developed by Intel and a consortium of industry partners. It’s a fantastic resource and, no matter which country you are based in, I would recommend attending to help you and your organisation identify where your breach priorities or security capabilities fall short of the industry and established best practices, which will enable you to make more informed decisions about where and how to invest future security spending.

 

The Cost Of Under-Investment In Security

There is, of course, a cost to not investing in security too. In Sweden, I have seen an example of the cost to a healthcare organization which suffered a ransomware attack. An infected file was opened from a webmail application while a doctor was connected to the hospital network. The malware began encrypting local files and those stored on the network, which included patient data from connected health centres outside of the hospital. Additionally, there was also a .txt file containing a ransom note.

 

Fortunately, the IT support team noticed the attack within 90 minutes and were able to successfully stop backups of the infected data and close down unauthorized access to the network. After many hours of work to rectify the breach, network access was restored some 22 hours after the initial attack. I estimate that the cost in IT resource time alone was somewhere in the region of 20,000 Swedish Krona, which equates to approximately $2,500 or €2,200. The cost in time lost by clinicians having to use workarounds and the potential loss had personal data got into the wrong hands would be multiples of this figure.

 

Learnings From Healthcare Security Breaches

I’m always keen to understand what lessons can be learned from security breaches such as that explained above, because only then can we start to win the battle against these cyberattacks and keep patient data safe and secure. Intel’s Security Maturity Model is a huge step forward in helping healthcare organisations better understand where they are today and where they need to go in order to mitigate the risks of a breach. This is why I’m delighted that our workshop at the end of May will bring together healthcare organisations and Intel security experts here in Sweden to share their knowledge.

 

- Contact the author: Johan Liden

- Security Workshop, Sweden, May 31st – June 1st: Register your interest

- Intel Health and Life Sciences: Security and Privacy

 

Consumer health was one of the big trends that came out of HIMSS 2016. Patients using wearable technology and smartphone apps to collect and send data to physicians is making a dramatic impact on how healthcare research is performed.

 

One area where this model is already moving forward is in Parkinson’s disease research. Patients battling this disease usually see their physicians every six to 12 months. By utilizing technology, patients can regularly collect data on their movements, send the information to the cloud for analysis, and be better prepared for their next appointment. This process provides more value for each interaction with the doctor and from what we see, the patients are excited to be able to contribute data and help researchers combat this disease.

 

In the above video clip, Chen Admati, advanced analytics manager at Intel, explains how consumer health platforms such as wearable technology are helping in Parkinson’s disease and shows how Intel is working to develop new algorithms to analyze important information. The hope is to take the value from this research model and translate it to other disease platforms to combat some of the most prevalent health challenges facing us today.

 

Watch the video and let us know what questions you have about wearables and consumer health.