A generative artificial intelligence (AI) diagnostic tool has been launched by Microsoft, which it claims is four times more accurate than experienced physicians and has the potential to ‘reshape healthcare’.
The Microsoft AI Diagnostic Orchestrator (MAI-DxO) was tested on 304 complex diagnostic cases published in the New England Journal of Medicine and achieved an 85% diagnostic accuracy rate, according to a recent study.
The accuracy of diagnoses by 21 physicians working in the US or UK was also tested in comparison, with a median of 12 years’ of experience. Of these, 17 were primary care physicians and four were in-hospital generalists. Their average diagnostic accuracy was 19.9%.
The research also looked at the cost of the tool, claiming it saved 20% in diagnostic costs compared to physicians.
The study, called Sequential Diagnosis with Language Models, which was not peer reviewed, was published on arXiv – a research sharing platform maintained by Cornell University.
Challenging the traditional structure of healthcare with AI
‘When doctors begin their careers, they face a key decision: should they become generalists, with broad knowledge across many medical areas, or specialists, with deep expertise in a narrow field? This division is necessary because medicine is too vast for any one person to master in full,’ the study said.
‘To manage this complexity, healthcare systems rely on collaboration: generalists and specialists work together in clinics and hospitals, combining their diverse and complementary knowledge and decision-making skills to provide patients with the comprehensive and effective care that they need.
‘Today, frontier AI language models are challenging this traditional structure. These advanced systems show remarkable versatility, demonstrating both broad and deep medical understanding, and the polymathic ability to reason across specialties. In effect, they combine the generalist’s range with specialists’ depth.
‘As a result, they significantly outperform individual physicians on complex diagnostic problems, such as those featured in the NEJM clinicopathological conference cases. Our findings highlight this impressive capability. Expecting any single doctor to master the full range of such cases is unrealistic.’
Eliminating or complementing doctors?
This study comes after Microsoft co-founder Bill Gates, suggested earlier this year on The Tonight Show that AI could mean doctors would not be needed ‘for most things’ within a decade.
However, Microsoft added that its MAI-DxO system is meant to ‘complement’ doctors and other healthcare professionals.
It said: ‘While AI is becoming a powerful tool in healthcare, our team of practising clinicians believes AI represents a complement to doctors and other health professionals.
‘While this technology is advancing rapidly, 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.
‘Clinical roles will, we believe, evolve with AI giving clinicians the ability to automate routine tasks, identify diseases earlier, personalise treatment plans, and potentially prevent some diseases altogether. For consumers, they will provide better tools for self-management and shared decision making.’
AI has a large role in the new NHS 10-year plan, which was published this week. This includes an AI scanning tool for NHS systems in order to flag safety issues in real time.
Last year, a white paper from the Microsoft company Nuance found that over a third of patients are in favour of clinicians using AI in consultations to improve documentation processes such as clinical letters.
Research by the Health Foundation also found that three quarters (75%) of the public support sharing some of their personal health data to aid the development of AI systems in the NHS.
A version of this article was originally published by our sister publication Healthcare Leader.
