xi's moments
Home | Technology

AI hospitals to revolutionize domestic healthcare services

By Cheng Yu in Hefei | China Daily | Updated: 2018-03-02 15:08

When a 71-year-old man, surnamed Hong, from Hefei, Anhui province, developed an unexpected headache, his grandson took him to a hospital to see a doctor.

Their first stop at the First Affiliated Hospital of the University of Science and Technology of China was the inquiry station to find out which department they needed to go to. They were greeted by a "nurse" who asked what symptoms Hong had and then told them to go to the Neurology Department and showed them the way on a 3-D map.

The "nurse", however, was not a person, but a robot using artificial intelligence.

The First Affiliated Hospital of USTC, also known as Anhui Provincial Hospital, is China's first intelligent hospital, where technology is helping to improve healthcare services and experiences.

The AI-enabled hospital, which launched the new program in August 2017, is a pilot between the hospital and iFlytek Co Ltd, a leading Chinese AI company.

"Magical!" Hong said on seeing the robot. "Our lives have indeed changed."

"I am a frequent hospital visitor, but this is the first time that I was served by a robot," Hong said.

The move is part of broader efforts to accelerate application of AI in healthcare, which has become necessary as China's aging population struggles to find enough highly experienced medical professionals.

Last year, the country unveiled a three-year plan to boost development of AI technologies and application of AI in several sectors including healthcare.

According to the plan, AI-enabled services robots will be widely used in China within three years, with robots that can help seniors and children being a priority.

In addition, the country aims to accelerate the use of AI-enabled systems to assist doctors with medical diagnosis and treatment.

At the intelligent hospital, technologies such as the robot guides are helping doctors and patients and revolutionizing how healthcare services are provided.

AI-enabled systems, for example, are helping doctors read CT images, speed up diagnosis and prevent misdiagnosis. Developed by iFlytek, the system features deep learning technology and has been programmed with thousands of labeled CT cases.

"While it usually takes three to five minutes for a doctor to read a CT case-which usually consists of hundreds of images-to make an initial diagnosis, it takes under one second for the AI system to do the same thing," said Tao Xiaodong, head of iFlytek Healthcare.

Tao said that the AI system can also learn from each case it reads so that its performance improves over time.

"Best of all, the system won't get tired," he added.

As of now, the AI-enabled system has helped to diagnose more than 47,000 CT images and accuracy for the detection of lung nodules, one of the indicators of potential lung cancer, has reached 94 percent. Another notable advance technology applied in the intelligent hospital is iFlytek's Speech EMR, which enables the transfer of doctors' advice into structured electronic medical records.

With a conversion accuracy rate of 95 percent, the product frees doctors from tedious documentation work so that they can focus on caring for patients. "Unlike other AI-based systems that focus on diagnosis and treatment for major diseases, iFlytek is exploring how to use AI to help general practitioners diagnose and treat common diseases," Tao said. "We are developing more AI applications to help doctors, patients and hospitals."

"In the coming years, we will help more hospitals become intelligent ones to make quality healthcare accessible to more people and help realize a 'Healthy China'," Tao said.

Global Edition
BACK TO THE TOP
Copyright 1995 - . All rights reserved. The content (including but not limited to text, photo, multimedia information, etc) published in this site belongs to China Daily Information Co (CDIC). Without written authorization from CDIC, such content shall not be republished or used in any form. Note: Browsers with 1024*768 or higher resolution are suggested for this site.
License for publishing multimedia online 0108263

Registration Number: 130349