Artificial intelligence can diagnose diseases just as accurately as trained doctors, research has shown

Artificial intelligence can identify diseases just as accurately as trained doctors, a major study claims.

Advertisements

Research shows that AI can detect a large number of conditions – from cancer to rare eye diseases – with the same precision as medical professionals.

Use the & # 39; s computer program & # 39; deep learning & # 39; to train themselves to recognize diseases by analyzing thousands of medical images.

It is based on data from previous health records to detect similarities in circumstances and to make an accurate diagnosis without human assistance.

An AI doctor can identify diseases as accurately as trained people, a large study claims (file image)

An AI doctor can identify diseases as accurately as trained people, a large study claims (file image)

Doctors who led the assessment claimed that AI has a & # 39; huge potential & # 39; has to improve the speed and accuracy of the diagnosis.

Advertisements

Researchers analyzed 14 studies comparing AI and physician diagnoses between January 2012 and June 2019.

Results showed that in-depth learning AI detected the disease correctly in 87 percent of cases – compared with 86 percent of doctors.

The ability of AI to exclude patients who had no disease was also slightly better – 93 percent compared to 91 percent of medical professionals.

However, the University Hospitals Birmingham NHS Foundation Trust team said that more tests were needed because there were only 14 tests left.

Chief author Professor Alastair Denniston said: & # 39; Diagnosis of disease using deep learning algorithms offers enormous potential.

MAT HANCOCK MATS FOR A & # 39; DIGITAL REVOLUTION & # 39;

Health Minister Matt Hancock has talked about his plans to include the NHS in a & # 39; digital revolution & # 39 ;.

& # 39; Our health service is on the eve of a technological revolution and our brilliant staff will be in the lead when it happens & he said in February.

Advertisements

& # 39; Technology will make the NHS the best in the world and I want everyone working in the health and care system to embrace it – from wearers to pathologists, surgeons to health care workers. & # 39;

Hancock has even accused the supermarket chain Tesco & # 39; more advanced and efficient & # 39; technology than the NHS.

In an important report, he also outlined plans for smart speakers such as Alexa from Amazon to use as & # 39; virtual medical coaches & # 39; to monitor patients at home.

Mr Hancock has insisted that e-mails become the main communication method for general practitioners and hospitals within two years, replacing letters.

Hospitals and general practitioners spend tens of millions of pounds a year on stamps, envelopes and paper, which he & # 39; obsolete technology & # 39; called.

Advertisements

The technically skilled MP said that 40 years after the first e-mail was sent, the NHS must accept that the technology will be in use for a long time.

He also prohibited hospitals from purchasing fax machines last year.

& # 39; Our review found that the diagnostic performance of deep learning models is equivalent to that of healthcare providers. & # 39;

Professor Denniston added: & # 39; But it is important to note that AI did not substantially exceed human diagnosis.

& # 39; From this exploratory meta-analysis, we cautiously state that the accuracy of deep learning algorithms is equal to healthcare professionals. & # 39;

Advertisements

He said that more studies with AI in real clinical situations are & # 39; needed.

Co-author Dr. Livia Faes of Moorfields Eye Hospital in London added: & Evidence about how AI algorithms will change patient outcomes should come from comparisons with alternative diagnostic tests in randomized controlled trials.

& # 39; Up to now, there have been hardly any studies where diagnostic decisions of an AI algorithm are performed to see what happens to outcomes that really matter to patients, such as timely treatment, hospital discharge time, or even chances of survival & # 39;

Dr. Peter Bannister, chairman of the care panel of the Institution of Engineering and Technology (IET), was skeptical about whether AI could be implemented on a large scale.

He said: & # 39; The application of artificial intelligence (AI) techniques to diagnostics continues to attract much attention, given the potential upside in terms of sensitivity, repeatability and throughput when applied to large, information-rich datasets, including medical images.

Advertisements

& # 39; So far, acceptance has been limited and there is skepticism as to whether these approaches can ever deliver a net patient benefit if implemented on a scale.

& # 39; This comprehensive study clearly illustrates what is possible, but also identifies the large evidence gap that almost all groups face that have tried to apply AI to diagnostics. & # 39;

The artificial intelligence in the study diagnosed a large number of serious conditions such as cancer, lung disease and heart problems.

A study showed that the deep learning AI correctly detected the disease in 87 percent of cases - compared to 86 percent by doctors (file image)

A study showed that the deep learning AI correctly detected the disease in 87 percent of cases - compared to 86 percent by doctors (file image)

A study showed that the deep learning AI correctly detected the disease in 87 percent of cases – compared to 86 percent by doctors (file image)

The review, published in The Lancet Digital Health, will address the major plans of the Secretary for Health and Social Care of a & # 39; digital revolution & # 39; for the NHS.

Matt Hancock has spoken openly about his hope to include AI, as well as digital medicine and robotics, in healthcare.

In February, Mr. Hancock said: “Our health care system is at the start of a technological revolution and our brilliant staff will be at the wheel when it happens.

WHAT IS DEEP LEARNING?

Deep learning is a form of machine learning that relates to algorithms that have a wide range of applications.

It is a field that is inspired by the human brain and focuses on building artificial neural networks.

Advertisements

It was originally formed on the basis of brain simulations and to make learning algorithms better and easier to use.

Processing large amounts of complex data then becomes much easier and allows researchers to trust algorithms to draw accurate conclusions based on the parameters that the researchers have set.

Existing task-specific algorithms are better for specific tasks and goals, but in-depth learning allows a broader scope of data collection.

& # 39; Technology will make the NHS the best in the world and I want everyone working in the health and care system to embrace it – from wearers to pathologists, surgeons to health care workers. & # 39;

However, the study has received cautious optimism from health experts across the country.

Advertisements

Paul Leeson, professor of cardiovascular medicine at Oxford University, said: & # 39; This article provides an overview of the current state of research that tests how well artificial intelligence identifies disease in a medical image compared to a doctor.

& # 39; The assessment has been carried out with great care, but real challenges have occurred to get a usable result.

& # 39; The authors had to combine findings from completely different medical problems and types of imaging, including research that was conducted at a very early stage of development.

Advertisements

& # 39; They also had to ask a very simple question, which is not really relevant to most healthcare AI applications.

& # 39; Clinicians are unlikely to use the results generated by AI separately, head-to-head with a computer, but rather combine the information from AI tools with other sources to decide how to best care for a patient.

Advertisements

& # 39; It is important that the work shows that a new research phase is needed, using more detailed investigations, to find the best ways to use artificial intelligence in healthcare. & # 39;

Richard Mitchell, professor of cybernetics at the University of Reading, said: “Great progress is being made in artificial intelligence, including the use of in-depth learning methods, and in some circumstances, such systems can perform better than humans.

& # 39; A problem with in-depth learning is that, unlike an expert system (a computer system that emulates the decision-making ability of a human expert), it is not easy to & # 39; explain why a certain result was achieved.

& # 39; There are some examples where a combination of human and artificial intelligence gives an even better result, and that might be the better route to take. & # 39;

. (TagsToTranslate) Dailymail (t) health

- Advertisement -