Microsoft says AI system better than doctors at diagnosing complex health conditions

A picture


Microsoft has revealed details of an artificial intelligence system that performs better than human doctors at complex health diagnoses, creating a “path to medical superintelligence”.The company’s AI unit, which is led by the British tech pioneer Mustafa Suleyman, has developed a system that imitates a panel of expert physicians tackling “diagnostically complex and intellectually demanding” cases.Microsoft said that when paired with OpenAI’s advanced o3 AI model, its approach “solved” more than eight of 10 case studies specially chosen for the diagnostic challenge.When those case studies were tried on practising physicians – who had no access to colleagues, textbooks or chatbots – the accuracy rate was two out of 10.Microsoft said it was also a cheaper option than using human doctors because it was more efficient at ordering tests.

Despite highlighting the potential cost savings from its research, Microsoft played down the job implications, saying it believed AI would complement doctors’ roles rather than replace them.“Their clinical roles are much broader than simply making a diagnosis.They need to navigate ambiguity and build trust with patients and their families in a way that AI isn’t set up to do,” the company wrote in a blogpost announcing the research, which is being submitted for peer review.However, using the slogan “path to medical superintelligence” raises the prospect of radical change in the healthcare market.While artificial general intelligence (AGI) refers to systems that match human cognitive abilities at any given task, superintelligence is an equally theoretical term referring to a system that exceeds human intellectual performance across the board.

Suleyman, the chief executive of Microsoft AI, told the Guardian the system would be operating perfectly within the next decade.“It’s pretty clear that we are on a path to these systems getting almost error-free in the next 5-10 years.It will be a massive weight off the shoulders of all health systems around the world,” he said.Explaining the rationale behind the research, Microsoft raised doubt over AI’s ability to score exceptionally well in the United States Medical Licensing Examination, a key test for obtaining a medical licence in the US.It said the multiple-choice tests favoured memorising answers over deep understanding of a subject, which could help “overstate” the competence of an AI model.

Microsoft said it was developing a system that, like a real-world clinician, takes step-by-step measures – such as asking specific questions and requesting diagnostic tests – to arrive at a final diagnosis.For instance, a patient with symptoms of a cough and fever may require blood tests and a chest X-ray before the doctor arrives at a diagnosis of pneumonia.The new Microsoft approach uses complex case studies from the New England Journal of Medicine (NEJM).Suleyman’s team transformed more than 300 of these studies into “interactive case challenges” that it used to test its approach.Microsoft’s approach used existing AI models, including those produced by ChatGPT’s developer, OpenAI, Mark Zuckerberg’s Meta, Anthropic, Elon Musk’s Grok and Google’s Gemini.

Microsoft then used a bespoke, agent-like AI system called a “diagnostic orchestrator” to work with a given model on what tests to order and what the diagnosis might be.The orchestrator in effect imitates a panel of physicians, which then comes up with the diagnosis.Microsoft said that when paired with OpenAI’s advanced o3 model, it “solved” more than eight of 10 NEJM case studies – compared with a two out of 10 success rate for human doctors.Microsoft said its approach was able to wield a “breadth and depth of expertise” that went beyond individual physicians because it could span multiple medical disciplines.It added: “Scaling this level of reasoning – and beyond – has the potential to reshape healthcare.

AI could empower patients to self-manage routine aspects of care and equip clinicians with advanced decision support for complex cases.”Microsoft acknowledged its work is not ready for clinical use.Further testing is needed on its “orchestrator” to assess its performance on more common symptoms, for instance.
societySee all
A picture

Most women in England and Wales have seen abusive male behaviour in past year, says poll

A majority of women have direct experience of violence or harassment, or know someone who has suffered it, in the last year, a poll has found.The poll finds little faith in the police or government to stem the tide of male violence, and most believe the problem has got worse.The survey was presented to a private meeting attended by police chiefs and police and crime commissioners just under three weeks ago.It was conducted by Zencity and based on almost 1,800 female respondents aged over 16 across England and Wales.The large scale and high frequency of violence against and harassment of women is something law enforcement and the government are trying to get a grip on

A picture

Health inequality is linked to gross disparities in wealth | Letters

Your article on health inequality (Britain’s ‘medieval’ health inequality is devastating NHS, experts say, 29 June) describes the laudable efforts of NHS agencies to tackle some of the acute health problems in poorer areas. However, the real problem is that the reason we have such disparities in health is that they are directly related to the gross disparities in wealth and income in this country.As Prof Michael Marmot and many others have demonstrated, some of the most important factors in determining health are social and economic. It is all very well for the NHS to make efforts to actively address the effects of social and economic deprivation in poor areas, but this is managing symptoms rather than the cause.It is no coincidence that the UK has some of the worst health outcomes of developed countries and also among the worst levels of inequality

A picture

The Vivienne died from cardio-respiratory arrest due to ketamine use, inquest finds

The drag artist known as The Vivienne died from misadventure after suffering cardio-respiratory arrest after taking ketamine, a coroner has ruled.James Lee Williams, 32, was found in the bath by a neighbour at home in Chorlton-by-Backford, Cheshire, on Sunday 5 January. The last time anyone had contact with Williams was two days earlier, a court was told, when a friend said it was evident the entertainer had taken ketamine.Five drug snap bags were found in The Vivienne’s property, including in a bedroom drawer and a bin in the bathroom, an inquest at Warrington coroner’s court heard on Monday.Although the performer had struggled with drugs in the past, Williams’s family told the hearing they should not be remembered for their use of ketamine and that drugs did not define the person they were

A picture

People having IVF should get time off work for appointments, say UK campaigners

People undergoing fertility treatment should have the legal right to take time off for their appointments, according to research that finds over a third have considered leaving their job due to the physical and emotional strain.The campaign group Fertility Matters At Work is calling for IVF to be recategorised as a medical procedure, rather than an elective treatment equivalent to cosmetic surgery, in guidance for employers under the Equality and Human Rights Commission (EHRC) code of practice.This would mean employers are no longer able to refuse time off for appointments, and would help tackle the stigma and lack of support that exists in many workplaces, the group says.Fertility Matters at Work has published a report based on a survey of more than 1,000 UK-based employees who have undergone fertility treatment. It found that nearly all (99%) had experienced it as a major life event that affected their mental wellbeing, while 87% reported anxiety or depression directly related to it, and 38% had left or considered leaving their job

A picture

NHS will use AI in warning system to catch potential safety scandals early

The NHS is to become the first health system in the world to use AI to analyse hospital databases and catch potential safety scandals early, the government has said.The Department of Health and Social Care said the technology will provide an early warning system which could detect patterns or trends and trigger urgent inspections. The scheme is part of the 10-year plan for the NHS that is due to be published by Wes Streeting this week.The government acknowledged the concern surrounding standards of patient care after “a spate of scandals including in mental health and maternity services”.Last week a national investigation into NHS maternity and neonatal services was announced by Streeting

A picture

Britain in 2025: sick man of Europe battling untreated illness crisis

The same 11 young women turn up around the clock at the emergency ward of Furness general hospital in Cumbria. The group are well known to staff, other services – and each other. Aged between 19 and 35, they have all led troubled lives. Some grew up in care, most need mental health support. All have fallen through society’s cracks and now gamble with their lives for a safe place to sleep