This paper reports new methods to examine oscillatory brain activity (or brain waves), which are thought to play a key role in how different brain regions communicate. These brain waves, which are altered in many diseases, can be thought of like walkie talkies tuning into the same frequency. Interestingly, one brain region can switch from one frequency to another to control what other brain regions it talks to. Current methods to assess brain waves average these signals over long time scales, obscuring dynamic changes in communication between brain regions. Combining signal processing and multiple machine learning methods, our new approach classifies oscillatory brain activity without averaging over time. This method provides a new approach to investigate oscillatory brain dynamics broadly without averaged across long time periods and therefore reveal sub second changes in how brain regions communicate with each other.