Researchers from the University of Madeira and different research institutes in Iceland, Finland and Spain developed an alternative view for sleep stability analysis. They evaluate sleep stability by creating an index based on the CAP A-phase characteristics, the A-Phase Index. 

The A-Phase Index is estimated with the use of machine learning. The researchers combine multiple artificial neural networks to assess the A-phase subtypes, their API, and the CAP cycles and rate. Their results show that this methodology is suitable to produce a fully automatic CAP scoring algorithm. Furthermore, a sleep stability profile can be acquired by studying when the A-phases occur and how long they last.

” This allows the visualization of sleep instability oscillations during the night and to observe when the brain creates periods of sustained stable sleep.”

The paper is published in SLEEP. You can read the full paper called “API: an Alternative View for Sleep Stability Analysis based on Automatic Detection of the A-Phases from the Cyclic Alternating Pattern” here.