论文部分内容阅读
With rapid revolution in technology and medicine, many new imaging tools are or will be available in clinical research and practice.The evidence-based evaluation of their benefits and appropriate use are necessary and impose huge statistical challenges.The essential difficulty is that the imaging techniques are primarily used as the starting point of medical intervention.Therefore, the performance of these techniques will depend on the follow-up medical treatment strategies and the final outcomes can take long time to observe and possibly be biased frown many confoun ding factors.The duration of study is important because the life-cycle of technology is much shorter than pharma ceutical interventions.In this talk, we will overview these statistical challenges, present statistical design for ima ging clinical trials to avoid/reduce the possible biases, and a new design for fast and initial assessment of imaging efficacy based on cross-sectional and short-term longitudinal follow-up data.