Infants receiving intensive care have measurements about their heart rate, blood pressure, saturation of gases in the blood, temperature and other vital signs recorded every second. The resulting time series contain a great deal of information about factors which cannot be examined directly, including the state of health of the infant and the operation of the monitoring equipment. Some temporal patterns are common and easily recognised, such as those associated with problems with monitoring equipment, while others are harder to interpret.
My PhD research involved formulating probabilistic dynamical models for the monitoring of premature babies, in order to make inferences about the state of the baby and the monitoring equipment.
PhD thesis, University of Edinburgh, 2007
List of other publications and media coverage
