# MedLog

## Docs

- [Attribute a downstream patient or system outcome to a prior inference](https://medlogprotocol.ai/api-reference/events/attribute-a-downstream-patient-or-system-outcome-to-a-prior-inference.md)
- [Record a new internal model artifact](https://medlogprotocol.ai/api-reference/events/record-a-new-internal-model-artifact.md)
- [Record content shown to a human user](https://medlogprotocol.ai/api-reference/events/record-content-shown-to-a-human-user.md)
- [Record explicit user feedback](https://medlogprotocol.ai/api-reference/events/record-explicit-user-feedback.md)
- [Record the beginning of an inference run](https://medlogprotocol.ai/api-reference/events/record-the-beginning-of-an-inference-run.md)
- [API reference](https://medlogprotocol.ai/api-reference/overview.md): The five write-once event endpoints that assemble a MedLog record.
- [Citation](https://medlogprotocol.ai/citation.md)
- [Deployments](https://medlogprotocol.ai/deployments/overview.md): Four clinical deployments monitored with MedLog.
- [Get involved](https://medlogprotocol.ai/get-involved.md): Share feedback, join the team, or pilot MedLog at your institution.
- [Integration patterns](https://medlogprotocol.ai/implementation/integration.md): Integrate MedLog at API gateways, LLM proxies, and agent frameworks — without modifying the models.
- [Interoperability](https://medlogprotocol.ai/implementation/interoperability.md): MedLog maps onto W3C PROV, OpenTelemetry, and FHIR without requiring a new software stack.
- [Low-resource settings](https://medlogprotocol.ai/implementation/low-resource.md): MedLog scales down to settings without full EHR infrastructure or continuous connectivity.
- [Storage and retention](https://medlogprotocol.ai/implementation/storage-retention.md): Logging volume depends on workflow design; retention policies keep it manageable.
- [What is MedLog?](https://medlogprotocol.ai/index.md): A protocol for event-level logging of medical AI.
- [Quickstart](https://medlogprotocol.ai/quickstart.md): Emit your first MedLog record by writing a stream of events to a collector.
- [Example record](https://medlogprotocol.ai/specification/example-record.md): An annotated, real MedLog record from the SEP-1 abstraction pilot in San Diego.
- [Header](https://medlogprotocol.ai/specification/header.md): Provenance information, execution context, and system metadata available at inference time.
- [Inputs](https://medlogprotocol.ai/specification/inputs.md): The input data provided to the model.
- [Internal artifacts](https://medlogprotocol.ai/specification/internal-artifacts.md): Artifacts generated during inference for technical researchers and MLOps engineers.
- [Model instance](https://medlogprotocol.ai/specification/model-instance.md): Stable identifiers of the AI model and version, with references to its model card and data sheet.
- [Outcomes](https://medlogprotocol.ai/specification/outcomes.md): Records of clinical actions or patient outcomes linked to the model recommendation.
- [Patient- or clinician-facing outputs](https://medlogprotocol.ai/specification/outputs.md): The outputs intended for human users.
- [Record schema overview](https://medlogprotocol.ai/specification/overview.md): Each MedLog record corresponds to one model invocation and contains nine fields.
- [Target identity](https://medlogprotocol.ai/specification/target-identity.md): A reference to the entity about which the model produces output.
- [User feedback](https://medlogprotocol.ai/specification/user-feedback.md): Any feedback provided by users, whether structured ratings or free-text comments.
- [User identity](https://medlogprotocol.ai/specification/user-identity.md): The technical process, service, or workflow that invokes the model call.
- [Team](https://medlogprotocol.ai/team.md): MedLog is developed by a global team across 51 institutions and 11 countries.

## OpenAPI Specs

- [openapi](https://medlogprotocol.ai/api-reference/openapi.yaml)