System studied electronic health records from 1.36 million pediatric visits in China
A team of Chinese and U.S. researchers have developed an artificial intelligence (AI) system that can "read" health records and give diagnoses as accurately as experienced human doctors.
In a paper published in Nature Medicine, the researchers said they have built an AI-based system that can automatically diagnose common pediatric diseases after processing the patient's symptoms, medical history and other clinical data.
To train and validate the system, the scientists used it to study and interpret electronic health records from more than 1.36 million pediatric visits to a major referral center in China.
The system shows a high level of accuracy comparable to that of experienced pediatricians and may aid doctors in diagnosing complex or rare diseases by providing more diagnostic predictions, according to the paper.
In addition, it may be useful to triage patients, which is highly important for areas short of medical resources.
"The study demonstrates that AI may help diagnose more diseases after systematically studying health records, but we still have much work to do," said Xia Huimin, co-author of the paper and a researcher with Guangzhou Women and Children's Medical Center. "We need a wider range of data to validate the system's level of accuracy."
"Besides, the collection and analysis of high-quality data are a long process that requires close cooperation between algorithm engineers, clinicians and epidemiologists," Xia added.
Click here for the complete issue.