| Popis: |
Sleep researchers classify critical neural events called spindles that are related to memory consolidation via scalp electroencephalography (EEG). Manual classification is time consuming and is susceptible to low inter-rater agreement. This could be addressed using an automated approach. The current study presents an optimized filter based and thresholding pipeline to set up a baseline for comparison to evaluate machine learning models using naïve features, such as raw signals, peak frequency, and dominant power. Filter based and thresholding pipelines allow us to formally define sleep spindles using signal processing but may miss examples most human scorers would agree were spindles. Machine learning methods, in theory should be able to approach human performance but they require a large quantity of scored data, proper feature representation, intensive feature engineering, and model selection. We evaluate both a pipeline based signal processing and machine learning with naïve features. We show that the machine learning models learned from our pipeline improve classification. An automated approach designed for the current data was applied to the DREAMS dataset. With one of the expert’s annotation as a gold standard, our pipeline yields an excellent sensitivity that is close to a second expert’s scores and with the advantage that it can classify spindles based on multiple channels if more channels are available. More importantly, our pipeline could be modified as a guide to aid manual annotation of sleep spindles based on multiple channels quickly (6-10 seconds for processing a 40-minute EEG recording) making it faster and more objective. |
