AI can analyze symptoms to accurately diagnose 87% of cases, 93% of cases recover, 86% and 91% of doctors.
The results of this study were recently published in Lancet Digital Health on the use of deep learning techniques in diagnosing diseases. This technique uses algorithms, computing power simulating human intelligence, thereby allowing computers to observe and analyze thousands of image data to make an diagnosis for each individual.
Hospitals around the world are adopting AI for healthcare.
"AI can detect diseases, from cancer to eye disease, but the ability to diagnose AI basically does not exceed human ability , " said Professor Alastair Denniston, University Hospital of Birmingham NHS Foundation Trust leads the research, said.
The global market for the application of artificial intelligence (AI) in health care is growing sharply, expected to increase from US $ 1.3 billion (in 2019) to US $ 10 billion (in 2024), according to state budget estimates. Morgan Stanley investment.
Hospitals around the world are using AI in healthcare, including Moorfields Eye Hospital, London. An algorithm written by the Google-owned DeepMind Research Center can help a doctor make a detailed diagnosis in about 30 seconds using an optical tomography (OCT) technique.
AI can also pinpoint some rare genetic disorders by analyzing images of patients' faces.
An application called DeepGestalt in three trials has shown its ability to outperform clinicians when detecting a range of symptoms.