Self-starting single control charts for multivariate processes: a comparison of methods

Random Forest Based Monitoring Approach to Detect Changes in Longitudinal Blood Pressure Data: A Retrospective Study

“…8 Other methods such as multivariate self-starting control charts can be used as an alternative to RTC. 6 However, these methods require a multi-conversion method where the feature parameters are estimated by the baseline. Even they work with small baseline data they perform better with larger datasets and require assumptions such as normality.…”

“…Many conventional methods were proposed to monitor longitudinal data. 6,18,19 Implementations to health data provide potential opportunities for disease management and public health surveillance. 6,7,19 These methods mostly rely on a control region showing "normal" estimated based on a baseline data.…”

“…6,18,19 Implementations to health data provide potential opportunities for disease management and public health surveillance. 6,7,19 These methods mostly rely on a control region showing "normal" estimated based on a baseline data. In practice, however, this assumptions can easily be violated, and this situation degrades the performance of the methods.…”

“…Process monitoring methods have been widely used in heath surveillance. 6,7 The daily rhythm of the measurements and the existence of a bivariate monitoring problem suggest that the appropriate method for monitoring is multivariate control charts. But the potential of these methods to detect undesired changes in BP limits the use of these methods as they require a long baseline data to estimate the limits of detection.…”