Medical Imaging Domain Ontology

FLOW ALGEBRA
TITLE: Medical Imaging Domain Ontology
DOMAIN: Domain Ontologies / Medical Domains / Medical Imaging
SOURCE: https://kirethresearch.org/booksai-core/ontologies/domain-ontologies/medical-domains/medical-imaging-domain-ontology.html
VERSION: 1.0
AUTHORS: Mark Anthony Higgins — Sky (Recursive AI)
STATUS: Companion formalisation; source-page specification not independently verified.

flow "MedicalImagingDomainOntology" v1.0 {
  ontology:
    id: "medical-imaging-domain-ontology"
    parent: "medical-domains"
    purpose: "Represent medical imaging studies, acquisition methods, anatomical observations, interpretations, measurement uncertainty and evidence-backed relationships."

  types:
    ImagingStudy: {id: string, modality: enum[XR, CT, MR, US, NM, PET, SPECT, other], acquired_at: datetime?, subject_pseudonym: string?, provenance: string}
    ImageSeries: {id: string, study_id: string, protocol: string?, frame_of_reference: string?, image_count: integer?}
    ImageInstance: {id: string, series_id: string, format: string, checksum: string?, pixel_spacing: string?, orientation: string?}
    AnatomicalStructure: {id: string, label: string, terminology_code: string?}
    ImagingFinding: {id: string, description: string, code: string?, anatomical_site: string?, confidence: number?, provenance: string}
    Measurement: {id: string, value: number, unit: string, uncertainty: number?, method: string?}
    Interpretation: {id: string, study_id: string, finding_ids: string[], interpreter_role: string, timestamp: datetime?, provenance: string}
    ImagingDevice: {id: string, modality: string, calibration_reference: string?}
    Relationship: {id: string, subject: string, predicate: string, object: string, confidence: number, evidence: string[], context: string?}

  predicates:
    contains_series(study: ImagingStudy, series: ImageSeries)
    contains_image(series: ImageSeries, image: ImageInstance)
    depicts(image: ImageInstance, anatomy: AnatomicalStructure)
    indicates_finding(image: ImageInstance, finding: ImagingFinding)
    located_at(finding: ImagingFinding, anatomy: AnatomicalStructure)
    measured_by(finding: ImagingFinding, measurement: Measurement)
    interpreted_in(finding: ImagingFinding, interpretation: Interpretation)
    acquired_with(study: ImagingStudy, device: ImagingDevice)
    supports(finding: ImagingFinding, interpretation: Interpretation)

  constraints:
    - "Every image must resolve to an existing series and every series to an existing study."
    - "Confidence values must be within [0,1]."
    - "Measurement values must include units and relevant uncertainty."
    - "Spatial measurements require pixel spacing, orientation and frame of reference where applicable."
    - "Image evidence does not alone establish a definitive clinical diagnosis."
    - "Do not merge patients, studies or findings using names or visual similarity alone."
    - "Preserve modality, acquisition protocol, laterality and anatomical context."
    - "Handle identifying imaging metadata under applicable privacy and access controls."

  inference:
    rule series_study_membership:
      if contains_series(study,series) and contains_image(series,image)
      then belongs_to_study(image,study)
      confidence: "structural provenance dependent"
    rule anatomical_evidence:
      if indicates_finding(image,finding) and depicts(image,anatomy) and validated_location(finding,anatomy)
      then evidence_of_location(finding,anatomy)
      confidence: "dependent on image quality and interpretation"
    rule measurement_trace:
      if measured_by(finding,m) and valid(m)
      then has_quantitative_evidence(finding,m)
      confidence: "retain measurement uncertainty"

  SEMANTIC_EQUIVALENCE:
    - "Map modalities and anatomy to versioned clinical terminologies where supported."
    - "Distinguish same finding, related finding and repeated observation."
    - "Preserve acquisition time, anatomical laterality and imaging frame."

  CONFIDENCE:
    range: [0,1]
    factors: [image_quality, acquisition_protocol, observer_agreement, measurement_uncertainty, provenance, validation_status]
    policy: "Do not increase certainty solely by chaining dependent imaging assertions."

  VALIDATION:
    - "Validate identifiers, study-series-image links, modality and metadata consistency."
    - "Check units, spatial frames, orientation, laterality and measurement uncertainty."
    - "Flag conflicting findings and unsupported diagnostic conclusions."
    - "Audit provenance, consent or legal basis, and access permissions as applicable."

  FLOW_HASH:
    algorithm: "SHA-256"
    input: "canonicalised ontology version, typed assertions, rules and provenance"
    note: "Compute hash at execution; no hash value is claimed here."

  UPDATE_MODEL:
    strategy: "versioned, provenance-preserving incremental update"
    steps: [ingest, deidentify_where_required, normalise, validate, compare, reconcile, recompute_inferences, publish]
    rollback: true

  AI_CONSUMPTION:
    representation: "typed medical imaging knowledge graph with traceable study-series-image and evidence relationships"
    outputs: [study_links, imaging_findings, anatomical_links, measurements, uncertainty_annotations, provenance_records]
    safeguards: [privacy_controls, no_unsupported_diagnosis, preserve_acquisition_context, human_clinical_review]
}

END FLOW ALGEBRA