In an average of 14 minutes, a mobile phone app can detect whether a hospital patient has a possibly deadly kidney disease. Through the technology created by Google’s DeepMind, alerts are sent directly to physicians and nurses. The app, called Streams, is looking for signs of acute kidney injury that kills about 100,000 individuals every year in the UK.
AKI includes sudden harm to the kidneys or reduced blood flow, usually the result of another disease. It’s not a kidney injury. Streams are looking for indications that the functioning of a patient’s kidney is deteriorating, which physicians say could save thousands of life. According to scientists at London’s Royal Free Hospital, patients may die, end up on dialysis or need a transplant without fast therapy. Experts think that if physicians are willing to intervene sooner, up to one in three AKI fatalities may be preventable.
Using restricted NHS technology, current techniques of identifying the disease can take several hours. Figures indicate that about one in five hospital patients suffer acute kidney injury (AKI), often caused by dehydration. The condition costs £1.2bn per year to the NHS. But there is a potential for the Streams app to save a lot of cash. Streams cut costs by approximately £2,000 per AKI hospital patient-from £11,772 to £9,761, according to outcomes released in Nature Digital Medicine.
An app trial discovered that it identified 96.7% of instances deemed an emergency-compared to 87.6% for present techniques. It operates by gathering information from IT devices already in the hospital, such as heart rate, blood pressure, and blood tests. The information is stored in one location along with the prior diagnoses of the patient, enabling clinicians to read through outcomes readily. Streams can send an urgent alert to assist the correct clinician, which would usually spend decreasing the time employees contacting each other.
For instance, if a blood test recorded elevated concentrations of a waste product called creatinine, which is usually filtered out by the kidneys, Streams would warn employees. The specialists assessed data from around 12,000 AKI alerts using Streams, involving a University College London team. The results discovered a’ important enhancement’ to quickly recognize acute kidney injury. But the retrieval time has not changed.
Dr Chris Streather, the Royal Free’s chief medical officer and deputy chief executive, said Steams results were ‘incredibly encouraging’.
He said: ‘Digital technology is the way forward for the NHS. ‘In the same way as we can receive transport and weather alerts on our mobile devices, doctors and nurses should benefit from tools which put potentially life-saving information directly into their hands. ‘In the coming months, we will be introducing the app to clinicians at Barnet Hospital as well as exploring the potential to develop solutions for other life-threatening conditions, like sepsis.’
To predict AKI, DeepMind also trialed an artificial intelligence system. In research, by analyzing up to 600,000 information points-such as blood tests, heart rate and blood pressure, the AI scheme effectively identified AKI two days early. The health and technology company’s algorithm was tested on records from the US Department of Veterans Affairs for more than 700,000 patients— the largest integrated healthcare system in the US.
It was able to detect 55.8 per cent of all inpatient episodes of AKI, according to findings published in Nature. More than 90 per cent of all acute kidney injuries that required subsequent administration of dialysis were detected. DeepMind are hoping to pilot the technology in UK hospitals within the next 12-18 months.
The researchers believe it could be extended to help detect other deadly conditions, including sepsis.Life-threatening sepsis is difficult to spot, and progresses rapidly. It was responsible for more than 350,344 hospital admissions in 2017/18. Dr Dom King, the health lead for DeepMind Health, said: ‘This progress represents potentially a very significant change in how medicine is practised and care is delivered.
‘A lot of care at the moment is very reactive and this represents the potential to re-move the needle to proactive, preventative care. And that requires a lot of careful thought.’ Dr King, previously a general surgeon, said the AI was able to look at thousands of signs of deterioration, but a nurse or doctor can only look at one or two.
He said: ‘The current alerts are very simple and rules-based and don’t really pick up the subtlety of those patients at the earlier signs of deterioration, so it [the AI] really is mindblowing for me as a doctor.’ Paul Leeson, professor of cardiovascular medicine, University of Oxford, said: ‘Trials are still needed to test whether this early warning is useful to doctors to improve patient care, without causing too many false alarms, or missing patients that the AI also overlooked.
‘However, this is another strong example of how AI appears to have the potential to augment delivery of healthcare.’ DeepMind’s technology has previously raised concerns about the privacy of NHS patients’ data.
DeepMind was bought by Google’s parent company Alphabet for £400million ($520m) in 2014. Based in London, it shares operations with the US-based Google Health unit, but Google states that patient data remains under DeepMind’s control and the move to Google does not affect anything.
An earlier deal between DeepMind and the Royal Free NHS Foundation Trust was found in 2017 to misuse patient data. When sharing information with DeepMind for work on Streams, The Royal Free did not comply with the Data Protection Act when it passed on personal information of around 1.6million patients, the Information Commissioner’s Office (ICO) found.