Skip to content

Skeletal region of Interest

Felix Engel edited this page May 9, 2017 · 3 revisions

A skeletal region of interest defines a part of the skeleton on which a specific observation can be made or which is needed for a specific type of analysis. Skeletal regions of interest can be defined in reference to various anatomical features like bones, substructures of bone (e.g. bony part of bone), bone surfaces etc.

Figure: General structure of skeletal regions of interest (cf. key to symbols in network graphs).

Composition of skeletal regions of interest

Skeletal regions of interest can scale on all levels, from skeletal regions covering several bones to individual bones and parts of bones. In order to secure exact definition and compatibility between data sets, the smallest building blocks of skeletal regions of interest are segments of skeletal elements, which define features of bones or teeth. These can be combined to form skeletal regions of interest that involve parts of several bones or teeth.

Segments of skeletal elements

Segments of skeletal elements define parts of anatomical structures. These can be anatomical features, arbitrarily defined sections or even anatomical structures in their entirety. Segments of sekeletal elements are defined as subclasses of rdfbones:SegmentOfSkeletalElement and are linked to elements from the Foundational Model of Anatomy through the object property obo-fma:regional_part_of.

Using the data property rdfbones:skeletalElementPortion, it can be specified for each segment of a skeletal element which portion of the referenced anatomical structure it represents. For instance, if an entire structure is represented, rdfbones:skeletalElementPortion will take the value 1.0. A Segment representing the surface of the left half of a mandible would have an rdfbones:skeletalElementPortion of 0.5. So rdfbones:skeletalElementPortion is defined for values between >0 to 1.0.

Combinations of skeletal segments

Previously defined segments of skeletal elements can be combined to define larger skeletal regions of interest. This is the only way to define skeletal regions of interest involving several bones. Combinations of skeletal segments are defined as subclasses of rdfbones:CombinationOfSkeletalSegments and are linked to the subclasses of rdfbones:SegmentOfSkeletalElement that form the combination through the object property obo:BFO_0000051 ('has part').

Both rdfbones:SegmentOfSkeletalElement and rdfbones:CombinationOfSkeletalSegments are subclasses of rdfbones:SkeletalRegionOfInterest.

Ontological integration of skeletal regions of interest

The concept of skeletal regions of interest is integrated into existing ontologies by defining rdfbones:SkeletalRegionOfInteres as a subclass of obo:FMA_63825 ('Anatomical zone'), defined as a "General anatomical term that is an arbitrary subdivision of an anatomical structure."

Application of skeletal regions of interest

Skeletal regions of interest mirror the natural anatomy as modelled by the Foundational Model of Anatomy. The reason for this duplication is that the requirements of scientific investigations not always follow the lines of natural anatomy. In order to make data queries more straightforward, all anatomy referred to in investigations is referenced as skeletal regions of interest, even if these only refer to natural anatomical structures.

The data property rdfbones:skeletalElementPortion can be used in several ways. First, all segments of skeletal elements with rdfbones:skeletalElementPortion = 1.0 can be treated as equivalents to the anatomical structures they represent. Also, the specified portion can be used to project data collected for segments of skeletal elements onto bones or other larger skeletal regions of interest.

Clone this wiki locally