Platform Capabilities

Everything You Need to Extract Lab Data

The only complete pipeline built for Spanish medical lab reports. From scanned PDF to FHIR R4 output with LOINC codes, UCUM unit normalization, and plausibility checks — in a single API call.

Core

Intelligent Extraction Engine

Primary AI engine with automatic fallback and optional advanced vision model for complex layouts. Automatic engine failover ensures maximum extraction accuracy.

Self-Hosted

100% Proprietary AI Models

Every AI model — OCR engines, vision models, and matching algorithms — is developed in-house. Your data never touches third-party APIs like OpenAI, Google, or AWS.

LOINC Code Mapping

8,400+ test name aliases mapped to standardized LOINC codes via an advanced matching algorithm: exact, prefix, substring, fuzzy, and spell correction.

Standards

FHIR R4 Output

Every API response can be returned as a standards-compliant FHIR R4 Bundle with DiagnosticReport and Observation resources.

Spanish-Native Processing

Purpose-built for Spanish lab reports. Handles European number formats (comma decimals), Spanish medical terminology, and regional lab conventions.

Smart Preprocessing

Automatic deskew, crop, binarization, and contrast enhancement. Table detection for structured extraction.

Plausibility Checks

Every extracted value is validated against physiologic plausibility ranges per LOINC code. Implausible results are flagged for review.

Unit Normalization (UCUM)

Automatic conversion to UCUM standard units with molecular-weight-specific conversion factors per analyte.

Enterprise Deployment

Docker + Kubernetes ready with HPA, PDB, NetworkPolicy, Prometheus metrics, and structured JSON logging.

Security

GDPR-Compliant Architecture

Stateless processing with no personal health data storage. All data is processed in-memory and returned immediately. Zero data retention.

Advanced

Advanced Vision Model

Optional AI Vision Model for complex or degraded documents that defeat traditional extraction engines.

Architecture

Intelligent Extraction Engine

Automatic engine failover ensures maximum extraction accuracy. Each tier activates only when the previous one cannot produce a confident result.

Tier 1 · Primary

Primary AI Engine

High-accuracy AI engine optimized for structured documents. Handles most lab report layouts with table detection.

Tier 2 · Fallback

Fallback Engine

Lightweight alternative that activates automatically when Engine 1 fails or returns low-confidence results.

Tier 3 · Advanced

Advanced Vision Model

AI Vision Model for complex or degraded documents that defeat traditional extraction engines.

All engines run within your infrastructure. Zero external API calls.

Standards

Standards-Compliant Output

Every API response can be returned as a fully valid FHIR R4 Bundle. DiagnosticReport and Observation resources are automatically generated with proper LOINC coding, UCUM units, and reference ranges.

FHIR R4 Bundle with proper resource references
LOINC-coded observations (8,400+ test aliases)
UCUM-standardized units with conversions
Reference ranges and plausibility flags

FHIR R4 Bundle — response.json

{  "resourceType": "Bundle",  "type": "collection",  "timestamp": "2026-02-24T14:30:00Z",  "entry": [    {      "resource": {        "resourceType": "DiagnosticReport",        "id": "report-001",        "status": "final",        "category": [          {            "coding": [              {                "system": "http://terminology.hl7.org/CodeSystem/v2-0074",                "code": "LAB",                "display": "Laboratory"              }            ]          }        ],        "code": {          "coding": [            {              "system": "http://loinc.org",              "code": "24323-8",              "display": "Comprehensive metabolic 2000 panel"            }          ]        },        "effectiveDateTime": "2026-02-24T10:00:00Z",        "result": [          {            "reference": "Observation/glucose-001"          }        ]      }    },    {      "resource": {        "resourceType": "Observation",        "id": "glucose-001",        "status": "final",        "category": [          {            "coding": [              {                "system": "http://terminology.hl7.org/CodeSystem/observation-category",                "code": "laboratory"              }            ]          }        ],        "code": {          "coding": [            {              "system": "http://loinc.org",              "code": "2345-7",              "display": "Glucose [Mass/volume] in Serum or Plasma"            }          ]        },        "valueQuantity": {          "value": 95,          "unit": "mg/dL",          "system": "http://unitsofmeasure.org",          "code": "mg/dL"        },        "referenceRange": [          {            "low": {              "value": 70,              "unit": "mg/dL"            },            "high": {              "value": 100,              "unit": "mg/dL"            }          }        ]      }    }  ]}

Ready to get started?

Read the documentation or reach out to discuss your integration needs.

Read the Docs Contact Us

Standards-Compliant Output

Every API response can be returned as a fully valid FHIR R4 Bundle. DiagnosticReport and Observation resources are automatically generated with proper LOINC coding, UCUM units, and reference ranges.

FHIR R4 Bundle with proper resource references

LOINC-coded observations (8,400+ test aliases)

UCUM-standardized units with conversions

Reference ranges and plausibility flags

{ "resourceType": "Bundle", "type": "collection", "timestamp": "2026-02-24T14:30:00Z", "entry": [ { "resource": { "resourceType": "DiagnosticReport", "id": "report-001", "status": "final", "category": [ { "coding": [ { "system": "http://terminology.hl7.org/CodeSystem/v2-0074", "code": "LAB", "display": "Laboratory" } ] } ], "code": { "coding": [ { "system": "http://loinc.org", "code": "24323-8", "display": "Comprehensive metabolic 2000 panel" } ] }, "effectiveDateTime": "2026-02-24T10:00:00Z", "result": [ { "reference": "Observation/glucose-001" } ] } }, { "resource": { "resourceType": "Observation", "id": "glucose-001", "status": "final", "category": [ { "coding": [ { "system": "http://terminology.hl7.org/CodeSystem/observation-category", "code": "laboratory" } ] } ], "code": { "coding": [ { "system": "http://loinc.org", "code": "2345-7", "display": "Glucose [Mass/volume] in Serum or Plasma" } ] }, "valueQuantity": { "value": 95, "unit": "mg/dL", "system": "http://unitsofmeasure.org", "code": "mg/dL" }, "referenceRange": [ { "low": { "value": 70, "unit": "mg/dL" }, "high": { "value": 100, "unit": "mg/dL" } } ] } } ]}