Yitu Healthcare’s new System can Now Diagnose Disease related to Paediatric more accurately

A new breakthrough by Chinese healthcare called Yitu has created a benchmark in Paediatric diagnosis. Yitu has successfully used natural processing language (NLP) to diagnose pediatric disease. The accuracy level of the test was as high as 97%. 

NLP is a branch of AI that assists computers to make sense of spoken and written language.

The new approach by Yitu has made its AI capabilities as good as doctors in terms of conducting diagnosis and reading electronic health.  The firm’s research has been showcased in Nature Medicine, a peer-reviewed medical journal.

The system which was used to diagnose was developed by using 101.6 million data points from over 1.36 million patients which visited over a period of 18 months. The system was bolstered by a set of data that combines the correct information and prior medical knowledge.

The system diagnosed patients who had pediatric diseases; which included 55 diagnostic codes for the diseases.  After conducting the tests on various patients, the system achieved robust performance common conditions as well as potentially life-threatening ones.

Apart from achieving an accuracy rate of 97% in identifying acute asthma exacerbations, the researchers also claim to achieve an accuracy rate of 93% in detecting diseases like roseola, meningitis, and varicella.

According to Ni Hao, president of Yitu Healthcare, the new achievement has proven that AI technology can assist doctors to work out on large data, as well as identify unclear and difficult medical cases.

The new breakthrough is another case of a medical AI which could potentially change the scenario in the healthcare sector.