In a described embodiment, a method of sleep stage detection is disclosed. The method comprises receiving a first vital sign feature 701a including a plurality of first feature values v1-v11 corresponding to respective ones of a plurality of epochs； and performing a first logistic regression operation 701a´ to calculate a first indication value 702a for each intermediate one of the epochs based on the corresponding first feature value v6 and those of preceding v1-v5 and succeeding v7-v11 ones of the epochs, the first indication value 702a being descriptive of a sleep stage of the corresponding intermediate epoch. A method of creating a model for logistic regression, a method of extracting a heart rate variability feature, a method of creating a model for extracting a heart rate variability feature, a method of deriving a medical dosage, a method of assessing a responsiveness to a medical dosage, a computer-readable medium, a computing device and a wearable device are also disclosed.
