YITU Healthcare deploys natural language processing for diagnoses

KUALA LUMPUR, Feb 13 (Bernama) – YITU Healthcare, through its natural language processing (NLP) system, can achieve high accuracy rates on par with doctors when reading electronic health records and generating patient diagnoses.

The result was published on Nature Medicine under ‘Evaluation and Accurate Diagnoses of Pediatric Diseases using Artificial Intelligence’, making it the first time a top medical journal has published research findings related to employing NLP technology in making clinical diagnoses.

The research was conducted by YITU Healthcare, in collaboration with Guangzhou Women and Children’s Medical Center and other research institutions.

The system has achieved high accuracy in many dangerous conditions including acute asthma exacerbation (97 per cent), bacterial meningitis (93 per cent), varicella (93 per cent), influenza (94 per cent) and roseola (93 per cent).

“It proves AI technology is able to assist doctors to deal with massive data and diagnose, as well as provide support, in uncertain and complex medical cases,” said YITU Healthcare chief executive, Ni Hao in a statement.

The system has great potential for clinical use, for example triage procedures. It generates predicted diagnoses with inputs of basic medical history, vital signs and physical examination of patients.

These predictions help doctors make better use of precious time. The system can also assist physicians to diagnose complex or rare conditions to avoid misjudgment or bias.

Most importantly, the system will provide high quality healthcare services as well as ease the tension of severe shortage of experienced paediatricians in China, benefiting both patient and doctor.

-- BERNAMA