Laboratory Results

Goal of Extraction

The Core Clinical Labs Library is a laboratory results-focused clinical data extraction tool designed to identify, extract, and ontologically map laboratory tests and results from unstructured clinical narratives and patient health record documentation. It extracts laboratory test names and quantitative values documented across clinical notes, laboratory sections, and test result summaries, standardizing representation through SNOMED CT concept mapping with associated confidence metrics.

Extraction Methodology

The extraction process operates at the sentence level, capturing each laboratory test with precise character-level locations (start and end positions) that enable direct traceability back to the “document”. Each extracted laboratory finding includes the test name, quantitative result value, measurement unit, and temporal context. When a patient’s laboratory test appears multiple times within a single document—whether from repeated orders, serial monitoring, or documentation across different note sections—those records are de-duplicated. We return the lab test instance in the document with the highest confidence if there are more than one lab test with the same SNOMED CT concept code and measurement value. The library performs extraction on a document-by-document basis; there is no automatic cross-document reconciliation or cross-patient data merging.

Note: A document is determined by our intelligent document splitting feature. Medical PDFs can contain a single document or multiple documents. A document’s type can provide clinical context important for the data extraction. This feature separates and categorizes an input PDF into individual documents by document type to prepare them for data extraction. An input PDF can remain one document or can be split into multiple documents through this process. A “document” in the context of this data extraction refers to the documents identified after the intelligent document splitting process.

Interpretation & Validation

Every extracted laboratory test is assigned a confidence score (0.0–1.0 scale) to quantify extraction certainty. This confidence score guides appropriate reliance on extracted data and supports informed clinical interpretation.

Clinical Considerations

Temporal context is critical for clinical decision-making. Understanding when laboratory tests were performed directly impacts interpretation of results and assessment of disease trajectory or treatment response. Serial laboratory results provide essential context for baseline establishment, trend analysis, and detection of clinically significant changes. Clinical judgment should be applied when interpreting extracted results, particularly for entries with lower confidence scores.

Target Concepts

Coding System: SNOMED CT

All extracted laboratory tests are mapped to SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms), with each test assigned a valid, active SNOMED concept identifier (6–18 digit numeric codes), along with its associated full SNOMED concept description. This structure provides both the standardized data needed for system integration and the contextual information needed for informed clinical interpretation.

Clinical Concepts Extracted

Laboratory tests are clinically documented quantitative measurements that have been performed and assessed that include a measurement value. This includes test results and measured values. Associated clinical attributes—including quantitative values, measurement units, and temporal context—are captured as they appear in source documentation.

We define laboratory tests with SNOMED concepts. Concepts that fall into the following clinical categories are included:

• Laboratory Procedures

The concept must be a specific, named laboratory test. A small number of overly broad or non-specific SNOMED concepts are excluded by design — for example, generic terms like “laboratory test (procedure) or “analysis (qualifier value)” add no specificity to a laboratory test.

The extracted concept must meet a confidence threshold. The extraction engine assigns a confidence score to each identified concept. In addition, this extraction only includes labs if they have a measurement value.

Extracted Fields

Labs Extraction Fields

Example of Extracted Output

• Note: This example demonstrates the content of the output, not the structure

Labs Illustrative Output Example

• Example JSONL object showing a typical extraction result

1
{
2
  "filename": "patient_1.pdf",
3
  "subdoc_id": "patient_1.pdf-0011",
4
  "manifest_index": 1
5
  "measure": "labs",
6
  "chartdate": "2022-07-05",
7
  "subject_id": "11110001",
8
  "processing_date": "2024-03-08T00:00:00",
9
  "data": [
10
    {
11
      "sentence": "ULAR: First and second sounds are heard. No murmurs appreciated. | ABDOMEN: Less distended. Bowel sounds are absent. | EXTREMITIES: He has 3+ pedal swelling. | NEUROLOGICAL: The patient is alert and oriented x3. Examination is nonfocal. |  | LABORATORY DATA: |  | ***White count*** is down from 20,000 to 12.5, hemoglobin is 12, hematocrit 37, and platelets 199,000. | Glucose is 157, BUN 14, creatinine 0.6, sodium is 131, potassium is 4.0, and CO2 is 31. |  | ASSESSMENT AND PLAN: |  | 1. Small bowel obstruction/paralytic ileus",
12
      "section_name": "PROCEDURES AND TREATMENT PROVIDED",
13
      "entity_text": "White count",
14
      "entity_start": 829,
15
      "entity_end": 840,
16
      "concept_id": "767002",
17
      "description": "White blood cell count (procedure)",
18
      "entity_confidence": 1.0,
19
      "temporality": "",
20
      "measurement_value": "12.5",
21
      "measurement_unit": "none"
22
    }
23
  ]
24
}