SLAM – A thin-client for interoperable annotation and biomedical signal handling

Abstract

Current approaches to biosignal handling in a medical environment increasingly rely on machine learning, which requires precise annotations. In multi-centric annotation scenarios, the diversity in annotation protocols leads to imprecise semantics of the annotations. This context shows the need for interoperable annotations, that can be reused with little cost across different protocols. The presented free and open-source software is an application for handling biosignals. It implements a novel method to add asynchronous and interoperable annotations with precise semantics. The annotations are stored in Allotrope Data Format (ADF), a structure based on the Hierarchical Data Format (HDF5). It allows for high performance I/O and aggregating of related patient data (e.g. different acquisition modalities). This research tool can be used solely in a web browser to facilitate integration in more complete environments.

Keywords

• Biosignals
• EEG
• Annotation
• Visualization
• Semantic web
• ADF