1. Introduction
This ontology defines concepts to describe the storage and exchange of biosignals,
regardless of their underlying file format, and provides terms for common biosignal
metadata.
A Recording contains individual signals, all of which have
been recorded together. Signals in a recording may be all from the one device (e.g. a multi-lead
ECG or EEG recording) or may be from different devices (e.g. a polysomnography recording).
Core attributes of Recordings are:
- subject
- Who or what is the subject of investigation.
- investigator
- Who made the recording.
- starttime
- When the recording began.
- duration
- The length of the recording.
- format
- How actual data is stored.
- dataset
- Where data is stored.
A Signal consists of a sequence of data points in some sampling dimension, usually time. Samples
may be uniformly spaced or irregular; the majority of biosignals are sampled at a uniform rate.
Core attributes are:
- label
- A descriptive label.
- units
- The physical units of the quantity measured.
- sensor
- The transducer or electrode used to capture the signal.
- source
- What produced the signal.
- filter
- Details of filtering applied prior to sampling.
- rate
- The sampling frequency.
- times
- An increasing sequence of sampling times.
An Event is something that has occurred in time, possibly for some duration. Sequences
of events, all describing aspects of the same quality, can be seen as a signal. Core attributes
of Events are:
- kind
- What the event represents.
- time
- When the event occurred.
- duration
- The length of the event.
Annotations are general notes, comments, or description of some qualitative measure,
and may be made about the whole, or some portion, of a Recording, Signal or Event. Core
attributes of Annotations are:
- target
- What the annotation describes.
- comment
- The text of the annotation.
- tag
- A semantic tag about the target.
- creator
- Who made the comment.
- created
- When it was made.
A Segment is some portion of a sampled entity, defined by an interval in the sampling
dimension, intended to be used as the target of an Annotation. Core attributes of a Segment
are:
- source
- The Signal or Recording which the Segment is part of.
- time
- The temporal interval over which the Segment is defined.
Other ontologies, besides this one, are used for some of the above attributes,
in particular classes and properties from DCMI Metadata Terms [1],
OWL-Time [2], the TimeLine Ontology [3],
and the Units of Measurement Expressions ontology [4].
3. BioSignalML ontology at a glance
An alphabetical index of BioSignalML terms, by class (concepts) and by property
(relationships, attributes), are given below. All the terms are hyperlinked
to their detailed description for quick reference.
Classes: Annotation, BP_Filter, Device, Electrode, ErrorTag, Event, EventType, Filter, HP_Filter, Instant, Interval, LP_Filter, Notch_Filter, Recording, RecordingGraph, SampleClock, Segment, SemanticTag, Sensor, Signal, SignalType, Simulation, Source, TemporalEntity, Transducer, UniformSignal, UnitOfMeasure,
Properties: clock, dataBits, dataset, eventType, filterFrequency, index, maxFrequency, maxValue, minFrequency, minValue, offset, period, preFilter, rate, recording, resolution, sensor, signalType, tag, time, uncertainty, units,
Individuals: BP, ECG, EEG, ErrorTAG,
This work is licensed under a
Creative Commons Attribute-ShareAlike 3.0 Unported License,
so you are granted to use, copy, distribute, transmit and adapt this work, but you must attribute the work to the
original authors and distribute the resulting work under similar terms. This copyright applies to the BioSignalML Ontology
and accompanying documentation in RDF. The BioSignalML Ontology uses W3C's RDF technology, an
open Web standard that can be freely used by anyone.