Unified Ontology of Human Function (UOHF)

Why a shared functional language is needed

Medicine can represent disease and treatment; anatomy represents structures; movement science represents tasks and training; rehabilitation and functional assessment represent performance and participation. These knowledge systems are valuable but not automatically composable. “Knee pain” is a symptom, “difficulty climbing stairs” is a task-level observation, “insufficient capacity” may be an assessed finding or a hypothesis, and “injury to a specific structure” is a different claim requiring specific evidence. Treating these as interchangeable creates misleading individual judgments.

UOHF attempts to keep different kinds of knowledge recognizable while allowing them to be connected through explicit human-function relationships.

The human-function concept and its essential distinctions

The fixed Chinese plain-language explanation is “身体被正常调用的能力。” A supporting English rendering is “The body's ability to be appropriately engaged and used.” Research and computation must still distinguish task demands, available capacities, actual engagement, observable performance, consequences, context, and time. Having a capacity does not guarantee its effective engagement in a particular setting. Reduced performance does not automatically imply that a capacity is absent; neither conclusion alone proves a disease.

UOHF provides semantic positions for these separate concepts so that observations, warranted inferences, and outstanding questions do not collapse into one undifferentiated record.

From people and contexts to evidence and feedback

The architecture is organized around connected professional dimensions: actors, contexts, anatomy, function, movement, symptoms, disease, assessment, recommendation, intervention, monitoring, learning, and feedback. Actors identify service roles and permissions; contexts specify tasks and conditions; anatomy, function, and movement link structures and real-world requirements; symptoms and disease remain distinct clinical categories; and assessment, recommendations, interventions, monitoring, learning, and feedback connect knowledge to practice and review.

These dimensions are an organizing design, not a claim that every layer and every terminology mapping has already been populated or released, and certainly not that an ontology can independently establish a clinical diagnosis.

Concepts, relations, mappings, and evidence require distinct governance

A human-function concept may have a preferred label, synonyms, Chinese and English forms, applicable contexts, and typed relationships. It may also carry mappings to external terminologies. Yet textual similarity is not clinical equivalence, and a code mapping is not evidence about a particular person. Concept and mapping versions, source provenance, evidence quality, and usage constraints must be traceable. Conflicts and uncertainties should be reviewed instead of silently overwritten.

Relevant external resources may include ICF, ICD, MeSH, and SNOMED CT where appropriate, licensed, and validated. UOHF is not proposed as a replacement for these existing classifications and terminologies; their intended purposes, versions, licensing terms, and mapping validity remain important.

UOHF in relation to HFWM and Human Function Intelligence

UOHF asks: What is being described, how does it relate to other concepts, and what semantic or evidence constraints apply? HFWM asks how the same individual's condition, demands, capacities, and engagement change over time. Human Function Intelligence investigates how to reason about the person, propose interpretations, and identify what evidence is still required. A shared ontology can help prevent a symptom from being promoted to disease, a model inference from becoming an observed fact, or a simulated learning exercise from becoming a real member-service record.

Why professional boundaries still matter

UOHF does not replace licensed clinical diagnosis, treatment, or safety decisions. Researchers, professionals, and AI tools must distinguish general knowledge from evidence about an actual individual and identify when a conclusion is not justified. A common language is valuable not because every discipline must speak identically, but because different disciplines can understand, test, and collaborate accurately.