TVGH kicks off a new era in big data research
Dr. Jaw-Ching Wu
Director, Department of Medical Research
Dr. Chun-Ying Wu
Chief, Translational Research Division
Big Data has become one of the most important parts in the development of artificial intelligence (AI). With the advantage of the National Health Insurance Research Database (NHIRD) providing detailed, high-quality and unified format data, Taiwanese researchers possessed the most promising opportunity in AI development. Director Jaw-Ching Wu from the Department of Medical Research, Taipei Veterans General Hospital (TVGH) addressed that thousands of research articles had been published based on NHIRD and led to altering many clinical practice guidelines and expert consensus in the past decade and further changed the patient care.
Although NHIRD has its advantages and contributions to clinical care, there are still limitations in research applications, such as inaccurate diagnosis coding, lack of healthcare records, lack of laboratory data and other images and pathological information. In order to compensate these defects, many medical centers in Taiwan have established in-hospital health big data centers to strengthen hospital infrastructure, facilitate clinical research and develop smart hospital. Under the leadership of the Superintendent Dr. Deh-Ming Chang and Director Jaw-Ching Wu, the TVGH Big Data Center was established within less than 9 months from the joint efforts by Dr. Chun-Ying Wu, Chief of Division of Translational Research, the Department of Information Management and many other experts in TVGH. It’s now open to all researchers in TVGH who are willing to conduct a big data clinical study.
There are two distinguishable features worth to address in TVGH Big Data Center. Firstly, the data structure is compatible with NHIRD data frame, and coding of laboratory data also adheres to international standard code. These enable the TVGH Big Data Center to integrate data seamlessly with the NHIRD and other medical centers’ big data. Not to mention, based on the funding from the Veterans Affairs Council (VAC), TVGH Big Data Center will work even closely with all other VACs’ healthcare services. Secondly, TVGH Big Data Center has collaborated with other well-known research institutes worldwide, including Stanford University (USA), Oxford University (UK), The Chinese University of Hong Kong and The University of New South Wales (AU) and published papers jointly in many high-reputational journals.
Dr. Chun-Ying Wu addressed that health big data researches could play a significant role in international cooperation including to integrate data from each country, to re-confirm study findings with their local context, and to accelerate the clinical applications. For instance, Dr. Wu’s research team analyzed 42,588 chronic hepatitis B (CHB) patients and found the pooled annual rate of HBsAg seroclearance was 1.02%. This is an excellent outcome of the joint work from TVGH and Stanford University researchers and is published in Gastroenterology journal in 2018.
Another study, collaborated with Oxford University research team, integrated 117 trials with total 32,852 patients in order to find the best choice for H. pylori eradication regimens. The study was published in Gut journal in 2018. Our finding identified the most effective regimen currently and evidenced the effectiveness of standard triple therapy was below therapeutic eradication rate in most of the countries. Longer duration was also associated with higher eradication rate.
The study published in Journal of Hepatology in 2018 is another successful story. Jointly worked by TVGH Dr. Chun-Ying Wu’s research team and researchers from Fu Jen Catholic University Hospital, Academia Sinica, and The Chinese University of Hong Kong, this study analyzed 23,851 CHB patients in Taiwan comparing with 19,321 CHB patients in Hong Kong who continuously receiving antiviral therapy and developed a prediction model for the risk of hepatocellular carcinoma (HCC) during the treatment. This study concluded the risk of hepatocellular carcinoma with a concordance rate around 75–80% during the first three years on oral antiviral therapy. TVGH currently have further integrated this prediction model with AI assistance. The HCC risk could be predicted more accurate in the near future.
Big data research also help to find the new uses for old drugs. The study is conducted by colleagues in Taipei and Taichung Veterans General Hospitals. They retrieved medical records from the NHIRD in Taiwan between 1998 and 2012 and screened the records of 204,507 patients with chronic HBV infection. The study finding pointed out that daily aspirin is significantly associated with a reduced risk for liver cancer in patients with HBV. This paper has been accepted to publish in the highest-ranking internal medicine journal, JAMA Internal Med in 2019.
The inauguration of TVGH Big Data Center will bring a new milestone in the big data research field in Taiwan. As Director Jaw-Ching Wu re-addressed, it is expected to contribute substantially in the health big data research and smart healthcare development in Taiwan. There is no doubt the TVGH Big Data Center will continuously collaborate with those outstanding research institutes around the world in extensive research area to benefit more to the patients.