The success of drug development and clinical research is heavily dependent on the quality of data, patience compliance, and statistical analysis. Traditional data collection via paper-pencil is labor- intensive and inconvenient, often leading to missing, noisy or invalid entries. There is always a challenge to monitor patients'medication compliance. When conducting analyses, traditional statistical methods lack of flexibility to make faster decisions. Seeing the ubiquitous application of wireless technologies and the quick emergence of Cloud Computing and Bayesian adaptive statistics in the pharmaceutical industry, we propose to develop a system called 'portable-devices based Data Solutions over the Cloud'(pDSC) for real-time data collection, scalable data storage, remote patient monitoring, and Bayesian adaptive clinical trial design. pDSC enables faster, cheaper, and effective clinical studies. Phase I of this project was highly successful. A prototype system was developed and tested by potential users. The business concept was highly acceptable for personalized data collection and analysis using the web-phone architecture. The feasibility of combing body area sensor networks with 3G wireless networks for collecting behavioral, psychosocial, geo-spatial and physiological data was fully verified. The protocols between iPhone and the web applications were developed and the method of high-performance Bayesian computation in the cloud was investigated. This Phase II proposal aims to fully develop the pDSC system to use smart phones and other portable devices for real-time data collection and patient monitoring. pDSC will be developed over the Windows Azure cloud platform with secure and scalable data storage mechanisms and interfaces to access other software tools for Bayesian adaptive design and interim analysis. The technical development will enable us to form a contracted research organization (CRO). It will offer cost-effective data services, fully complying with FDA guidelines, to help academic and pharmaceutical industry clients develop medicines to meet the goal of personalized treatment and patient-centered care.
Quality of data and analysis is crucial for the success of any clinical research. The proposed solution based on wireless and Internet technologies and Bayesian adaptive methods will greatly enlarge the capacity and cost-effectiveness of data solutions in developing pharmaceutical products. It has great potential to meet the needs of next generation clinical research and healthcare delivery featured by personalized medicine, patient-centered service, and prevention-oriented care.
| Zhang, Xiaoshuai; Xue, Fuzhong; Liu, Hong et al. (2014) Integrative Bayesian variable selection with gene-based informative priors for genome-wide association studies. BMC Genet 15:130 |
| Zhang, Xiaoshuai; Yang, Xiaowei; Yuan, Zhongshang et al. (2013) A PLSPM-based test statistic for detecting gene-gene co-association in genome-wide association study with case-control design. PLoS One 8:e62129 |
