Cyber PSG is a software framework with semi-automatic classification of biosignals, optimized for the analysis of sleep data. The system is hardware independent and allows the implementation of an interface for the reception of various signals from various types of devices.
Using advanced methods, indications are extracted (feature extraction) that are needed for the automatic classification of individual segments of records into hierarchically-arranged tree clusters. The system then performs the classification of the recorded data into a tree structure of hierarchically-arranged classes (clusters) and offers the expert key parts of the record recommended for evaluation.
The system enables the combination of manual classification of the record by an expert with a high degree of automation, based on the evaluation of the similarity of the signals implemented in the form of hierarchical clustering.
The main advantages of the developed approach are:
- Significant acceleration of PSG recording classification while maintaining full expert control over the classification process
- Increasing the objectivity of record evaluation compared to purely manual classification
- Increase in classification accuracy (compared to fully automatic classifiers)
- The possibility of use regardless of the type of measurement (after optimizing the system for the given type of task) or the hardware used (the system is not tied to a specific device manufacturer)