Whenever we report a test result or estimate some parameter of interest, we are interested in quality of the results. Uncertainty analysis provides a systematic approach to estimating the quality of the results from a test where the test can be anticipated or already completed. A number of methods in different fields related to biomedical instrumentation and signal processing can be redefined to include uncertainty and uncertainty quantification such as design of experiments, decision making and risk analysis, estimation and data fusion.
We will show how to present uncertainty using statistical intervals and how to propagate uncertainties from the input to the output of the mathematical model. In addition, we will describe at a very high level how to estimate statistical intervals of the estimates when the processing is performed in real time. We will also show how to handle problems where the models are non-linear and non-Gaussian.
Miodrag Bolic (M’04–SM’08) received his PhD degree in electrical engineering from Stony Brook University, USA, in 2004. Since 2004 he has been with the University of Ottawa, Canada where he is an associate professor with the School of Electrical Engineering and Computer Science. Dr. Bolic’s current research includes biomedical signal processing and instrumentation. His current projects are related to development of contactless monitoring systems using radars and cameras as well as development of single-arm ECG wearable devices. He published about 60 journal papers, 4 book chapters and edited one book.