Shuang Wang, Ph.D.

CTO, Novo Vivo and Assistant Professor, the School of Informatics, Computing, and Engineering at Indiana University

王爽

现任美国印第安纳大学，信息计算工程学院助理教授, 美国NovoVivo Inc，CTO。曾任加利福尼亚大学圣地亚哥分校（University of California，San Diego）助理教授。其主要研究方向包含生物医疗信息，数据隐私与安全，大数据分析，机器学习， 高性能计算，数据压缩。他领导开发的独特的基于SGX硬件技术的大规模基因安全数据分析模型2016年获得了Intel的杰出贡献奖。

Biomedical data privacy standards and research: challenges, strategies and technologies

生物医疗数据隐私标准及其研究：挑战、应对策略与技术

Healthcare big data is fostering a huge market, bringing technological advancement or value transformation. At the same time, there are new challenges that come with it: when individual’s healthcare data are brought together, it may become a "magic mirror", telling the master behind it, such as who are you, where are you, or even more secrets about you. Since you share a part of the common genomes with your immediate family member, the disclosure of your personal medical and genetic information may also have a negative impact on the privacy of them. Therefore, it is imperative to protect the privacy of biomedical big data. Sufficient protection is necessary to safeguard patient privacy and to increase public trust in healthcare research. In this presentation, we will first review the existing challenges of healthcare data privacy. Then, we will discuss both policies and technological solutions to regulate and safeguard healthcare data privacy, respectively. Regarding policy, we will cover the Health Insurance Portability and Accountability Act (HIPAA) in US, General Data Protection Regulation (GDPR) in EU, and Network Security Law of the People's Republic of China. Regarding technological solutions, we will discuss data de-identification method under HIPAA and the risk assessment strategies of the HIPAA safe harbor rule based on large-scale Chinese healthcare data. Then, we will introduce homomorphic encryption, federated learning and differential privacy technologies for healthcare data privacy protection. More specifically, we will focus on the studies of privacy-preserving federated learning technologies in both structure and non-structure data under horizontal and/or vertical partition paradigms. Through this presentation, audience are able to become familiar with both regulatory and technological progresses in healthcare data privacy protection.