©wladimir1804 -
stock.adobe.com
t
ACCIDENT & EMERGENCY
Embracing the digital revolution
Themovement towards a greater reliance uponmachine learning, artificial intelligence (AI) and digital technology as diagnostic tools is one that is poised tohaveasignificantimpactonUKTrusts. TheClinicalServices Journal explores the diverseways organisations aremaking themost of these new technologies.
Recent figures obtained from the NHS show that medical negligence claims arising from misdiagnosis are on the rise. In 2016/2017, the NHS paid out £1.024 billion for diagnoses which were either too late, were missed, or were just plain wrong. What’s more, 1534 successful claims were brought against the NHS, 28% more than the year before and the highest number in six years of data provided in a Freedom of Information (FOI) request. Despite this, there is little to suggest that the abilities/good intentions of the medical professionals have changed. Josh Hughes, head of complex injury at Bolt Burdon Kemp believes one possible explanation is the increasing resource and funding pressures placed upon an already stretched NHS. “Doctors and nurses are being asked to do more than ever, with the same or less resource, against a backdrop of unprecedented demand from an aging population,” said Hughes. “As a result, the usage of machine learning and artificial technology (AI) as a diagnostic tool is an exciting one.”
By way of example, researchers at an Oxford hospital have developed AI that can diagnose scans for heart disease and lung cancer. The technology aims to tackle the alarming fact that 12,000 in every 60,000 heart scans each year are misdiagnosed. At present, cardiologists diagnose heart disease by scanning the heart to monitor the heart rate. However, using this method, one in five people are misdiagnosed. They either have a heart attack or have had unnecessary surgery. The AI system developed by researchers at Oxford Radcliffe Hospital in Oxford is designed to capture the details that physicians cannot see and make more accurate diagnoses. The system, called Ultromics, is trained
to identify potential problems from a cardiac scan of 1000 patients treated in the past seven years and to provide information on whether these patients will have heart problems later on. Paul Leeson, the system’s researcher and cardiologist, said the data shows that
JUNE 2018
Given the sheer volume of data it will be capableof processing,AI could identify complex patterns and use them to predict or treat
illnesses far more efficiently than we can now. Josh Hughes, Bolt Burdon Kemp
Ultromics outperforms the experts and improves the diagnosis of coronary heart disease by more than 90%.
Echocardiography uses sound waves to image the heart. Because of its safety profile, low cost and ease of deployment, echo is the most widely used cardiac imaging test. Heart disease affects almost 50% of people over the age of 40 – making it the biggest killer globally. However, making a diagnosis from echo relies on experienced clinicians having to make qualitative judgements based on only a fraction of the data that is potentially available to them from a typical scan. The Ultromics technology extracts more
than 80,000 data points from a single echocardiogram image to overcome subjectivity and increase diagnostic accuracy of coronary heart disease from 80% to >90%. Its applications analyse all these data points using machine-learning techniques and have the potential to yield new, objective and highly diagnostic metrics about cardiac function and disease. Anticipating the launch of the first stand-alone stress echocardiography product for cardiologists in 2019, the researchers also have a future product pipeline growing out of one of the world’s largest research databases of echo images.
WWW.CLINICALSERVICESJOURNAL.COM I 75
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80
