Seoul Collaborates with Hospitals to Support Digital Health Startups
The city government of Seoul, South Korea, has joined forces with the Ministry of Health and Welfare to aid health and medical startups in leveraging medical and health data from hospitals. The Ministry of Health and Welfare plans to select seven startups to facilitate connections with 43 hospitals that currently provide de-identified, anonymized data for the development of new digital technologies, including AI-driven solutions.
Moreover, the selected startups will receive up to 50 million won (approximately $35,000) in business funding, along with expert consultation in medical data analysis and research conduct. This initiative aims to foster innovation in the healthcare sector and promote the development of cutting-edge technologies that can enhance patient care and outcomes.
Hong Kong’s eHealth App Enhances Access to Pathology Reports
Users of Hong Kong’s eHealth mobile health app can now access pathology reports through the Investigations function. The Health Bureau has made pathology reports available for viewing 14 days after their release by the Hospital Authority, Department of Health, and private healthcare providers.
Since its launch in 2021, the eHealth app has expanded its services to include access to nine types of electronic health records, with plans to incorporate additional features such as radiology images, Chinese medicine prescription records, and dental check-up details. This expansion aligns with the five-year eHealth+ plan, which aims to streamline healthcare data sharing and service delivery through the mobile app.
National Cancer Center Utilizes NLP to Formalize Pathology Data
The National Cancer Center in South Korea has successfully employed natural language processing (NLP) models to extract key information from free-entry, semi-structured breast pathology reports. The research team at the center optimized existing NLP models, including BERT-basic, BioBERT, and ClinicalBERT, using 1,215 breast cancer pathology reports.
By fine-tuning these NLP models, the researchers achieved a high accuracy rate of 96% and above in formalizing the pathology reports. This breakthrough has alleviated the burden on hospitals to manually convert unstructured medical data into a structured format, which is crucial for cancer research. With recent advancements in government regulations allowing for the use of NLP in data conversion, healthcare institutions can now streamline data processing and enhance research efforts in the field of oncology.
