Monday, March 24, 2014 - 08:00
Skaggs School of Pharmacy and Pharmaceutical Sciences, San Diego, CA
Biomedical Science has been undergoing a significant transformation in the past decade towards data-driven approaches, thanks to the rapid advances in computing technologies, which make a large amount of biomedical data (e.g., human DNA sequences, biomedical images, etc.) increasingly available. Such a "big data" science heavily relies on analysis, integration and mining of the data on an unprecedented scale. The technical challenge that comes from this demand has been further complicated by the need to preserve the privacy of the data, which oftentimes carries patients' sensitive information (e.g., their health records, genomic sequences, etc.). Actually, such a privacy concern has already made data owners (e.g., medical centers or genomic researchers) cautious about sharing biomedical data, e.g., to impose complicated procedures to review the applications for data access. This development unfortunately makes these data less accessible to researchers. To overcome this barrier and facilitate convenient access to biomedical data, effective privacy preserving techniques need to be developed to strike a balance between data sharing and privacy protection.
To answer the urgent call for practical protection of biomedical data, we propose Critical Assessment of Data Privacy and Protection (CADPP) as a community effort to facilitate development of new privacy-preserving dissemination techniques for biomedical data, and evaluate those techniques on real-world data, which includes both human genome data and those combined with clinical demographics. After the competition, a workshop will be organized to discuss the evaluation results, and the teams whose approaches achieve best results in each track will be invited to present their methods.