Streaming the right information to the right clinician at the right time

With our Streams app, we’re helping transform the way nurses and doctors prevent avoidable patient deterioration in NHS hospitals. Too many lives are lost because clinicians don’t have real-time information about who urgently needs their care. One in ten NHS in-patients suffers some kind of avoidable harm, and 40% of people in intensive care could have been prevented from ending up there. We think technology can help.

Our vision is of an NHS where every patient receives the right care from the right clinician at the right time, made possible through cutting-edge mobile technology that pushes patient alerts to nurses and doctors, enables them to securely assign and communicate about clinical tasks, and gives them all the information they need to make the right diagnoses and decisions. 

We’re working closely with our partner trusts to make this vision a reality. Streams' technology will speed up the time to alert nurses and doctors to patients in need down to a few seconds. And by freeing clinicians from juggling multiple pager, desktop-based, and paper systems, Streams will help redirect significant amounts of time from admin and towards direct patient care. 

We’ve built Streams in close partnership with NHS clinicians, who know exactly what they need, and are holding ourselves to the highest standards of security and oversight. That includes building new infrastructure that will give our partner hospitals an additional real-time and fully proven mechanism to check how we’re processing data. We're also working with digital product studio ustwo to bring the very best of user-centric design techniques to the field.

In time, we hope that this can help unlock the next wave of innovation in the NHS. The infrastructure that powers Streams is built on state-of-the-art open and interoperable standards (known as FHIR). This infrastructure will allow our partner trusts to have other developers build new services that integrate more easily with their systems, helping them bring new innovations to the clinical frontlines.

fullscreen fullscreen_mobile

What Streams does

Based on what clinicians have said they need, our partner Trusts will have the ability to implement the following:

  1. Results viewing. Providing nurses and doctors with quick and easy access to all of their patient’s test results.
  2. Alerts. Notifying nurses and doctors as soon as observations or test results indicate a potential problem with a patient, helping to provide much faster care for widespread problems including acute kidney injury, sepsis and more.
  3. Patient medical history. Presenting clinicians with key facts about the patient’s medical history such as previous diagnoses, procedures and admissions, so clinicians have the fullest understanding of their condition in deciding the right care. 
  4. Teamwork. Replacing the current outdated system of allocating tasks amongst clinical teams by phone, paper, pagers and fax machines, by allowing nurses and doctors to request actions from other members of the team, such as looking at test results or checking on a patient.
  5. Messaging. Enabling frontline staff to communicate simply and securely with one another about the care they’re providing. 

Bringing these features together in one app could save millions of hours every year that would otherwise be wasted by sharing information using the current, out-of-date technology, which in turn could save millions of pounds for the NHS. 

play Streams Features

Clinicians play an invaluable role in helping us ensure our products are as clinically effective as they can be. We are looking for nurses and doctors working in hospitals who would be interested in testing and providing feedback on the Streams application at the DeepMind office in Kings Cross. Click here if you’d like to find out about getting involved.

Our Partners

The Royal Free London NHS Foundation Trust

Since late 2015 we’ve been working with the Royal Free London NHS Foundation Trust on Streams, starting out with a version of the app focused solely on acute kidney injury (AKI). AKI is a contributing factor in up to 20% of emergency hospital admissions, and is linked to 40,000 deaths in the UK every year – a quarter of which NHS England estimates are preventable.

The partnership began when kidney specialists approached us to ask if technology could help them to detect AKI in its early stages. Every minute counts for patients who develop AKI: it can result in minor loss of kidney function all the way through to dialysis, transplant, and even death.

Michael Wise, a patient who developed the most serious type of AKI while in hospital, describes his experience of undergoing dialysis and receiving a kidney transplant, and the impact of his illness on his family, here:

play Acute Kidney Injury – a patient's journey

To help clinicians to detect AKI at an early stage, the app automatically alerts them as soon as a blood test indicates a possible AKI, and provides them with all the historical patient information they need to make the right diagnosis.

Streams was rolled out at the Royal Free in early 2017 and has shown early signs of improving patient care. By the end of February, more than 26 nurses and doctors at the Royal Free were using Streams, with the app alerting clinicians to an average of 11 patients at risk of acute kidney injury per day. Some of the nursing staff using Streams have estimated that it has been saving them up to two hours every day, which means they can spend more time face-to-face with patients.

Over the coming months Streams will roll out to more clinicians at the Royal Free and a formal service evaluation will be carried out to measure the overall impact Streams is having.

We’re also pleased to have entered into a longer-term partnership with the Royal Free that will see us extend Streams to cover illnesses other than AKI, including conditions such as sepsis and other organ failure, as well as implementing additional features such as clinical task management. 

We’re proactively publishing our agreements with the Royal Free with minimal redactions, including our master services agreement (covering the partnership as a whole) and our information processing agreement (covering how patient data is processed).

Some reactions to our new partnership with the Royal Free:

Professor Keith McNeil, NHS chief clinical information officer:
"The health and care system stands poised to harness the power of information and technology to substantially improve the care we provide to patients and to promote world class health outcomes.

“We cannot do this alone and we need to work with world leading partners, and I am delighted that the Royal Free are going to be able to use the brilliant technology and innovation offered by a partner of the stature of DeepMind to help deliver better, safer care to their patients. I commend them for their initiative and wish them every success with the venture.”

David Myers, President of the Royal Free Kidney Patients Association: 
“As a kidney patient for more than 25 years I am really excited about the potential of the Streams AKI app. This seems to me something that will revolutionise the speed with which AKI patients will be cared for. I think this system could not only save lives but could improve patients chances of living longer whilst reducing their stay in hospital. This must not only benefit patients by ensuring that they receive the correct care immediately but also should reduce the cost of caring for patients if AKI incidents are recognised at an early stage. I am happy that patient data is safe.”

Professor Jane Dacre, President of the Royal College of Physicians:
“The Royal Free London NHS Trust and DeepMind’s new partnership […] offers exciting possibilities for improved care for patients and more efficient systems for physicians to work with. It is encouraging to see the NHS harnessing technology to make a difference at the bedside and the clear focus on delivering real differences to patients’ lives and health.  Whenever we develop new ways of using patient data, it is essential that safeguards are in place for appropriateness and confidentiality, but with these we should embrace the opportunity to improve healthcare quality and reduce the burdens of bureaucracy on clinicians so they can focus on their patients.”


Imperial College Healthcare NHS Trust

In December 2016 we agreed a five year partnership with Imperial College Healthcare NHS Trust to help it bring on board the latest technology in delivering patient care.

Over the last two years the Trust has moved from paper to electronic patient records, and mobile technology is the natural next stage of this work. By giving clinicians access to cutting-edge healthcare apps that link to electronic patient records, they’ll be able to access information on the move, react quickly in response to changing patient needs, and ultimately provide even better care. 

We’ll be working with the Trust to deploy our clinical app, Streams, which supports clinicians in caring for patients at risk of deterioration, particularly with conditions where early intervention can make all the difference. Like breaking news alerts on a mobile phone, the technology will notify nurses and doctors immediately when test results show a patient is at risk of becoming seriously ill. It will also enable clinicians at Imperial to securely assign and communicate about clinical tasks, and give them the information they need to make decisions on care.

As in our partnership with the Royal Free, we’re also implementing state-of-the-art open and interoperable standards, via what’s known as a FHIR API. This will allow the Trust to easily, securely and consistently integrate other apps that could improve care, whether developed by third parties or innovators within the Trust. Our partners at Imperial are excited about the potential for a wide range of apps to improve care, and we’re delighted to be working with them to make this possible.

We’re proactively publishing our agreements with Imperial with minimal redactions, including our master services agreement (covering the partnership as a whole) and our information processing agreement (covering how patient data is processed).

Some frequently asked questions

Streams is a secure clinical app used at the Royal Free to help nurses and doctors treat hospital in-patients who are at risk of becoming seriously unwell. It automatically analyses test results, and if it picks up signs of deterioration then it sends an immediate mobile alert with all the relevant information to the right nurse or doctor, so they can come to the bedside right away. The first area we are working in is acute kidney injury, which is a sudden deterioration in kidney function that causes them not to work properly. The early feedback from nurses, doctors and patients has been very positive, and you can read more about one of the patients who was helped in the Evening Standard newspaper, here.

Some people have asked questions about how DeepMind’s Streams app works, and so we’ve answered some of the most common questions below. 

Streams and The Royal Free

  • Why does DeepMind process the data of patients from the Royal Free?

    For Streams to work, it has to securely process patient identifiable data. Otherwise, it would be impossible for the app to help nurses and doctors know who needs treatment, and what’s wrong with them. Like any company working with the NHS to process patient data, DeepMind has to follow a strict set of rules, and ensure the data is only used on the instructions of the hospital.

  • What does your work with the Royal Free involve?

    The Royal Free uses DeepMind’s Streams technology to help doctors and nurses check for signs of a kidney condition called Acute Kidney Injury (AKI).

    AKI refers to a sudden loss of kidney function, which can be fatal. It contributes in up to 20% of emergency hospital admissions and is linked to 40,000 deaths in the UK every year, a quarter of which NHS England estimates are preventable.

    Streams is helping to solve this problem. The app processes blood test results and other information to generate instant alerts to kidney doctors and nurses, warning them that a patient might be suffering from AKI, so they can come to the bedside right away and offer specialist treatment.

    We’re already seeing very exciting signs that instant alerts are improving outcomes for patients at the Royal Free. Nurses at the hospital also tell us that having access to patient information in the app is saving them hours every day, giving them much more time to spend with patients. You can read about this here


  • Who approved this partnership?

    The partnership was proposed by some of the UK’s leading kidney experts based at the Royal Free, and was approved by the Board of the Royal Free. The Streams app itself is officially registered with the national Medicines and Healthcare products Regulatory Agency (MHRA). DeepMind’s data storage infrastructure and practices have also been audited by NHS Digital (formerly known as HSCIC) and found to meet the highest standards.

  • Why does the Streams app process a wider range of data, not just blood tests?

    The amount and type of data Streams processes is determined by what clinicians at the Royal Free say is necessary to provide the best care to patients. 

    Doctors diagnose AKI by analysing blood test results that show whether the kidneys are operating normally. But because other medical factors affect how the kidneys operate, those results differ from person to person.

    Pregnancy, for example, alters kidney function. That means pregnant women have different test results, so a blood test result that’s healthy for one woman could be dangerously high for another. Including that information in the app, alongside instant alerts, means doctors and nurses can take the most accurate decisions – it gives them a more accurate picture about whether a result means that a patient is doing well, or if it means they need immediate medical help. 

    The same is true for other types of patient information. Doctors and nurses need Streams to tell them what other medical conditions or procedures patients have had, for example a previous kidney transplant, or if they’re diabetic. That information is shown in the app because doctors and nurses need it in order to provide the safest and most appropriate care for the specific patient in front of them. 

  • Why does Streams process data for patients who haven’t had blood tests?

    One of the most important benefits of Streams is that it presents the right patient data to the right clinicians quickly, integrating data from a number of different electronic systems in one place. Clinicians need to have quick access to a wide range of information about patients’ previous medical history and laboratory results (including blood tests) to make safe and accurate diagnoses. For example, if blood tests pick up a patient suffering sudden AKI, it’s important for clinicians to know if the patient has had a previous kidney transplant. To deliver fast and reliable alerts to clinicians through mobile devices, that relevant patient information must be stored in advance ready to be sent to a doctor or nurse at the first sign of a problem in the blood tests.

  • Why does Streams process historical data too?

    Historical information is vitally important for doctors and nurses when they are deciding the best treatment. One of the reasons AKI is challenging to treat is because the blood test result that is used to detect it has to be compared to earlier results when the kidneys were operating normally. Without that extra context, it’s impossible for clinicians to tell if a result reflects normal kidney function or is dangerously higher.

    Other historical data is also relevant in planning treatment for AKI, so the app needs to include that information, too. Patients who have had kidney problems in the past, for example, are much more likely to develop very serious forms of AKI, so it’s important that kidney doctors are made aware of this as quickly as possible so they can give patients the most appropriate treatment. The same is true for whether a patient has recently had emergency surgery, or if they’ve had heart disease. 

  • Why can’t Streams just process data from patients who are in the hospital and known to be unwell?

    AKI can affect patients for a wide variety of reasons - including patients who develop AKI as a consequence of another procedure, such as a hip replacement, or because of another medical condition, such as pneumonia or sepsis. That means it’s very difficult to predict exactly which patients will develop AKI. In addition, AKI quite often develops without patients showing any symptoms, so it’s also incredibly hard to isolate just those patients who have it from those who do not.

    Because it can strike anyone in the hospital, doctors and nurses at the Royal Free told us that Streams needed to process information about all patients accessing services. If it did not, then patients could develop AKI without it being detected early by the Streams app.

    For example, in one day in February, Streams alerted clinicians to 11 patients at risk of AKI, ranging from someone suffering from cancer, to an A&E patient, to someone in the maternity ward.

  • Does Streams use an unusual amount of data?

    No. There are many other companies that provide vital IT services to the NHS, and which process many millions of patient records. By comparison with these companies, DeepMind’s data processing is relatively small. All these companies, like DeepMind, have rightly to stick to strict rules to keep this data safe and ensure it’s only used for the agreed purposes.

  • Did the first 2015 agreement between the Royal Free and DeepMind (which has since been superseded in 2016 by a new agreement) go beyond what was necessary for acute kidney injury?

    No. The initial agreement between DeepMind and the Royal Free, signed in 2015, made clear that DeepMind was processing data strictly under the instructions of the Royal Free (which remains the case today, under our revised contracts). These instructions were to process data needed for the detection and treatment of acute kidney injury. 

    The amount and type of data Streams processes is determined by what clinicians at the Royal Free say is necessary to provide the best care to patients.

    Doctors diagnose AKI by analysing blood test results that show whether the kidneys are operating normally. But because other medical factors affect how the kidneys operate, those results differ from person to person.

    Pregnancy, for example, alters kidney function. That means pregnant women have different test results, so a blood test result that’s healthy for one woman could be dangerously high for another. Including that information in the app, alongside instant alerts, means doctors and nurses can take the most accurate decisions – it gives them a more accurate picture about whether a result means that a patient is doing well, or if it means they need immediate medical help. 

    The same is true for other types of patient information. Doctors and nurses use Streams to tell them what other medical conditions or procedures patients have had, for example a previous kidney transplant, or if they’re diabetic. That information is shown in the app because doctors and nurses need it in order to provide the safest and most appropriate care for a specific patient. 

    All this patient information is already being processed by other systems used in the hospital. DeepMind’s role is simply to process this information so it can be presented to nurses and doctors via a secure mobile app, helping them to deliver better and faster care.

  • Does Streams use AI? If not, then why is an AI company like DeepMind building it?

    Streams doesn’t currently use AI. Right now, we’re simply focusing on getting the right test results to the right nurse or doctor via a secure mobile app. This is an essential first step before any more advanced technology like AI can be introduced.

    Of course, this first step can also bring enormous patient benefits. We’re already hearing stories of people whose care has been helped by Streams, and some nurses are saying that the technology is already saving them two hours each day. You can read more here

  • What are you doing to protect patient data?

    Before processing any patient data, we undertook a major project to design an information and data security architecture of the highest quality, working in partnership with clinical and security experts to analyse and mitigate risks, as well as conducting on-site risk assessments with partners.

    We completed a self-assessment of our information governance toolkit and assessed our compliance to be at Level 3, the highest level possible across each of their requirements (NHS Digital’s assessments verified that we had reached Level 2, the highest level at which NHS Digital assesses organisations as standard).

    All data access is logged, and subject to review by our Trust partners. In addition, our data and technical infrastructure arrangements can be openly scrutinised by DeepMind Health’s nine Independent Reviewers

    We’re also building on this further to give our partner hospitals an additional real-time and fully proven mechanism to check how we're processing data. That will allow our partners to continuously verify that our systems are working as they should, and that data is only being used as it should be. This is being built by one of the world’s leading security engineers, Ben Laurie, co-founder of the OpenSSL project which enables encrypted connections to websites around the world (familiar to millions through the padlock in their browser bars).

  • What’s different in the new partnership with the Royal Free compared with your previous agreement with the Trust?

    Following our work on kidney injury, we wanted to explore whether Streams can be used to improve patient care in other ways, to help make the Royal Free one of the safest hospitals in the world.

    In November 2016 we signed a new agreement with the Trust, and will be expanding our alerting technology to a range of other potentially fatal conditions, including sepsis and organ failure. Alerting doctors and nurses to patients who need their attention in seconds rather than hours could dramatically improve patient safety.

    We’re also planning to include the task management features of Streams at the Royal Free, allowing them to view and update records and assign clinical tasks to each other from their mobile, which we hope will eliminate the need for doctors to shuffle through paper and receive pager alerts.

    By knitting this system together into an infrastructure based on open standards, other medical innovators will also be able to develop their own technologies for the Royal Free.

    Finally, we’re also building an unprecedented level of data security with our audit infrastructure, which will allow the Royal Free to verify exactly when and by whom patient information is accessed, with no possibility of falsification or tampering.

  • What guarantees can DeepMind give that patient data isn’t being misused, and that it isn’t being shared with Google?

    DeepMind will only ever process patient data on the instructions of our hospital partners, and we will always keep that data safe and secure within England. We will never link any of this data to any Google products or services.

    These protections are built into our legal contracts (which we publish on our website) as well as the laws and regulations that govern patient data.

    Our use of data is fully logged and audited, so our partner hospitals can always check who has had access to patient data and for what purpose. 

  • Are patients giving consent for you to use their data?

    When patients go into hospital, doctors and nurses don’t ask patients whether they consent to letting them use their medical information as part of their treatment. If they did, that would slow down the treatment process enormously, stopping doctors and nurses from treating patients as quickly as possible. 

    The same is true when doctors and nurses use IT systems, which are an essential part of modern healthcare. They can reasonably assume that patients want them to use IT systems to pull up their medical records, process their test results and help them provide patients with treatment. 

For Streams to work, it has to securely process patient identifiable data. Otherwise, it would be impossible for the app to help nurses and doctors know who needs treatment, and what’s wrong with them. Like any company working with the NHS to process patient data, DeepMind has to follow a strict set of rules, and ensure the data is only used on the instructions of the hospital.

General

  • What’s in it for DeepMind? Will the company make money from patient data?

    We do not and will not make money from patient data - we don’t sell it, we don’t use it for advertising, we don’t commercialise it in any way. Our business model is simple: we will charge hospitals a fee for providing them with a useful service that improves patient care - like other health technology providers do. 

  • Where is this headed long-term? Will technology like Streams replace nurses and doctors one day?

    Absolutely not. Technology like Streams is a tool to help doctors and nurses do their jobs better – it can never substitute for a qualified medical professional or a human’s diagnosis.

  • Can patients get involved in your work?

    Yes. Patients have valuable expertise to offer in how they’re treated, and health outcomes are better when clinicians and patients make decisions together. We ran our first patient participation event at our offices in September 2016 and we’ll be running others across the UK this year, where we hope patients will be able to give us their feedback and help set our priorities. You can read more about this here.

  • What is a FHIR API?

    Patient information is a vital part of clinical care – doctors and nurses treating you in hospital need to be able to access your complete medical history to help diagnose what is wrong with you and work out the best course of treatment.

    Streams helps bring that information together and allow doctors to access it securely and instantly on smartphones when they need it. But to make this happen, Streams and other clinical apps and systems need to use a shared computer language.

    We have committed to building our infrastructure on open and interoperable standards, namely the FHIR API (Fast Healthcare Interoperability Resources Application Programming Interface) created by the Health Level Seven International (HL7) health-care standards organisation. Not only will this ensure the data we process is in a modern infrastructure, but it will help to develop common information processing standards that other technologists and clinicians can also use to build their apps and other software (subject to those third parties seeking all the appropriate approvals).

    FHIR is being rapidly adopted by a wide array of health and social care providers internationally (eg. Boston Childrens’), and is supported in the UK by NHS Digital, NHS England and the INTEROPen group.

  • What data does Streams process for Imperial College Healthcare NHS Trust?

    We haven’t yet started processing data for Imperial College Healthcare NHS Trust, but expect to start in 2017 as our technologies go through clinical safety testing followed by a phased deployment.

  • How can I opt out of my data being processed for Streams?

    Our partner hospitals are in control of which patient data is processed, and so patients need to discuss opting out with their hospital directly.

  • How long do you have access to patient data?

    Our agreements with the Royal Free and Imperial last until 2021, before which date the data will be passed back to the Trusts or destroyed unless the agreement is extended.

We do not and will not make money from patient data - we don’t sell it, we don’t use it for advertising, we don’t commercialise it in any way. Our business model is simple: we will charge hospitals a fee for providing them with a useful service that improves patient care - like other health technology providers do.