Keynote Speaker

Chen Wen-Xi

Prof. Wen-Xi Chen

Title: From Gold Mining to Data Mining – Big Data Analytics and Lifelong Healthcare

170 years ago, California Gold Rush boosted the “Gold Mining” industry, and eventually boomed “Silicon Valley”. Nowadays, we are creating 2.5 quintillion bytes of data every day. Big Data come from everywhere: sensors used to gather climate information, posts to social media sites, digital pictures and videos, purchase transaction records, mobile phone signals, and personal information. Using state of the art atomic-scale magnetic memory technologies, researchers have demonstrated the possibility to store one bit of data by only 12 atoms which is potentially 100 times denser than today’s hard disk drive technology.

Changes in quantity must bring on changes in quality — Big Data are more than simply a matter of size. It paves a way for new and emerging realms, opens the door to a substantial world of opportunities for us to explore, to make our work more agile, our life cozier, and to answer questions that were previously considered beyond our reach.

In this talk, a pioneer project, “Challenge to 100 years of age” which was funded by Japanese government and involved more than 600 residents in West Aizu village since 1994, will be reviewed; the latest advancements of Big Data analytics in lifelong healthcare domain worldwide will be briefly outlined; some of our study outcomes on system development and data analysis in cooperation with nursing homes and local hospitals in recent years will be introduced. These studies utilized multiple vital signs, spanned several years, and covered various subjects, including pregnant women, healthy subjects, chronic patients and elders. Data analysis was conducted on different temporal bases, such as daily, weekly, monthly and seasonal. The results reveal long-term dynamics in health condition change of patients and healthy subjects, unveil elder and youngster daily behaviors, contribute to better understanding of maternal cardiac changes during pregnancy and after delivery, and find insights into more sensitive vital parameters from Big Data analytics. Three application scenarios are envisaged: real-time detection of emergent situations; long-term tracking of health condition changes; personal optimization of living environment setting. It is believed that, in the near future, a traditional house will be transformed into a fascinating at-home facility for lifelong healthcare and wellness management to provide emergency surveillance, to facilitate daily health monitoring, and to support higher quality of life.

Short biography: 
Wenxi Chen received B.S. degree and M.S. degree in biomedical electronics from Zhejiang University, Hangzhou, China in 1983 and in 1986, respectively. He received Ph.D. degree in biomedical instrumentation from Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, Japan in 2001. He is currently a professor of Biomedical Information Technology Laboratory (BIT Lab), and Director of Research Center for Advanced Information Science and Technology (CAIST), the University of Aizu, Aizu-Wakamatsu, Japan. He have participated more than 15 major R&D projects funded by Japanese ministries, Fukushima prefecture and industrial sectors with totally about 600 million JPY since 1998. These activities produced about 40 patents, 200 scientific papers and several book chapters. Some academic outcomes have been commercialized by relevant companies, and reported by various mass media. He has several memberships in professional societies such as Japanese Society for Medical and Biological Engineering, IEEE Engineering in Medicine and Biology Society (EMBS), Japan Association for Clinical Monitoring, Institute of Basal Body Temperature Measurement Promotion. He has hosted, co-organized, and served for many international conferences, domestic conferences, and professional committees. His current research interests are focusing on several aspects, such as developing various modalities to monitor diverse vital signs under daily life environment, performing comprehensive physiological interpretation of multifarious long-term data over multiple time scales to reveal statistical links and causalities among diseases, healthy aging and lifestyle as well as how they interact with various exogenous factors such as meteorological, environmental, geographical facets in temporal/spatial domains. In the end, these outcomes are expected to boost fostering a new interdiscipline “Metrology of Health” or “Healthology” based on a holistic view of health and disease.


Prof. Wen-Lian Hsu

Title: Explainable AI - An example from biological literature mining

In this work, we introduce an interpretable machine learning (ML) method, the statistical principle-based approach (SPBA), for biomedical named entity and relation recognition (BNER & BR). SPBA is a hybrid ML and knowledge-based approach that can identify biomedical named entities (BNEs) and relations (BR) by exploiting fine-grained linguistic knowledge represented as human-readable semantic concepts and patterns. SPBA automatically groups similar patterns into clusters and picks a representative pattern from each cluster as a principle. SPBA uses a flexible alignment matching algorithm that allows for insertion and deletion; bigrams consisting of inserted words and their neighboring words are weighed using logistic regression according to whether they give rise to true or false positive instances.

Brief Introduction:
Wen-Lian Hsu (F'06) is currently a Distinguished Research Fellow of the Institute of Information Science, Academia Sinica, Taiwan. He received a B.S. in Mathematics from National Taiwan University, and a Ph.D. in operations research from Cornell University in 1980. He was a tenured associate professor in Northwestern University before joining the Institute of Information Science in Academia Sinica as a research fellow in 1989. Dr. Hsu's earlier contribution was on graph algorithms and he has applied similar techniques to tackle computational problems in biology and natural language. In 1993, he developed a Chinese input software, GOING, which has since revolutionized Chinese input on computer. He later applied similar semantic analysis techniques to question answering system and chatbot. Dr. Hsu is particularly interested in applying natural language processing techniques to understanding DNA sequences as well as protein sequences, structures and functions and also to biological literature mining. Dr. Hsu received the Outstanding Research Award from the National Science Council in 1991, 1994, 1996, the first K. T. Li Research Breakthrough Award in 1999, the IEEE Fellow in 2006, and the Teco Award in 2008. He was the president of the Artificial Intelligence Society in Taiwan from 2001 to 2002 and the president of the Computational Linguistic Society of Taiwan from 2011 to 2012.