AI beats Chinese doctors in competition diagnosing brain tumours

A Chinese AI system has diagnosed brain tumours faster and more accurately than a team of doctors in competitions designed to showcase the new technology.

During two rounds of tests in Beijing pitting top Chinese physicians against a new, specifically designed AI system called ‘BioMind AI’, the ‘machines’ won on both occasions.

The BioMind AI system has been developed by the Artificial Intelligence Research Centre for Neurological Disorders at the Beijing Tiantan Hospital alongside a research team from the Capital Medical University.

In the competition, AI made a correct diagnosis in 87 per cent of 225 cases in about 15 minutes whilst a team of fifteen senior clinicians achieved only 66 per cent accuracy.

The AI also gave correct predictions in 83 per cent of brain haematoma expansion cases (the most devastating form of stroke) outperforming the doctors again who achieved a 63 percent accuracy.

The results led the head of the radiology department at Tiantan Hospital, a leading centre for neurology and neurosurgery, to clarify the results for the human physicians were even better than average accuracy in ordinary hospitals.

To train the AI, developers fed it tens of thousands of images of nervous system-related diseases that the Tiantan Hospital has archived over the past 10 years, making it capable of diagnosing common neurological diseases such as meningioma and glioma with an accuracy rate of over 90 percent, comparable to that of a senior doctor.

All the cases were real but the AI had not been fed the images in training, the organisers said.

But the hospital was quick to dismiss ideas that AI could replace doctors, saying the contest to pit doctors against technology was a learning tool for its staff.

Wang Yongiun, executive vice president of the hospital, said he personally did not care very much who won.

“I hope through this competition, doctors can experience the power of artificial intelligence. This is especially so for some doctors who are skeptical about artificial intelligence. I hope they can further understand AI and eliminate their fears toward it,” said Wang.

One of the doctors who took part in the test said that she welcomed AI. Dr Lin Yi called it a “friend” not a threat which would reduce workload for doctors and push them to keep learning and improving their skills.

‘Preliminary Results’

Bian Xiuwu, an academician with the Chinese Academy of Science and a member of the competition’s jury, said the AI would only serve as an assistance to doctors in giving preliminary results. Dr Paul Parizel, former President of the European Society of Radiology and another member of the jury agreed that AI would not replace doctor but instead act as a guide.

Andrew Bird from GSM said: This is an exciting development and good news for patients and the medical profession alike and it shows the huge potential that healthcare has in utilising and developing AI to improve results for patients.

“Much time in hospitals is spent waiting for doctors to look at results, particularly in departments like Accident and Emergency, because they are the majority of the clinicians who have the ability to diagnose. In an organisation like the NHS, with its increasing demands on front line services and ever present financial problems, embracing AI could be an important solution in addressing the future strategy of the NHS.”

China has introduced a series of plans in developing AI applications in recent years.

In 2017, the State Council issued a development plan on the new generation of Artificial Intelligence and the Ministry of Industry and Information Technology also issued the “Three-Year Action Plan for Promoting the Development of a New Generation of Artificial Intelligence (2018-2020).”

The Action Plan proposed developing medical image-assisted diagnostic systems to support medicine in various fields.