Clinical TempEval - Temporal Information Extraction from Clinical Notes
Extraction of temporal information from clinical text, identifying time expressions, event mentions, and their temporal relations. Based on SemEval-2016 Task 12 (Clinical TempEval).
Configuration Fileconfig.yaml
# Clinical TempEval - Temporal Information Extraction from Clinical Notes
# Based on Bethard et al., SemEval 2016
# Paper: https://aclanthology.org/S16-1165/
# Dataset: http://alt.qcri.org/semeval2016/task12/
#
# This task asks annotators to identify time expressions and event
# mentions in clinical text, and classify the temporal relation between
# events and times.
#
# Span Labels:
# - Time Expression: Date, time, duration, or frequency expression
# - Event Mention: Clinical event (procedure, condition, medication, etc.)
#
# Temporal Relation Labels:
# - Before: The event occurred before the time reference
# - After: The event occurred after the time reference
# - Overlap: The event overlaps with the time reference
# - Before-Overlap: The event began before and overlaps the time reference
# - Contains: The time reference contains the event
annotation_task_name: "Clinical TempEval - Temporal Extraction"
task_dir: "."
data_files:
- sample-data.json
item_properties:
id_key: "id"
text_key: "text"
output_annotation_dir: "annotation_output/"
output_annotation_format: "json"
port: 8000
server_name: localhost
annotation_schemes:
- annotation_type: span
name: temporal_entities
description: "Highlight time expressions and clinical event mentions in the text."
labels:
- "Time Expression"
- "Event Mention"
- annotation_type: radio
name: temporal_relation
description: "What is the temporal relation between the primary event and time reference?"
labels:
- "Before"
- "After"
- "Overlap"
- "Before-Overlap"
- "Contains"
keyboard_shortcuts:
"Before": "1"
"After": "2"
"Overlap": "3"
"Before-Overlap": "4"
"Contains": "5"
tooltips:
"Before": "The event occurred before the time reference"
"After": "The event occurred after the time reference"
"Overlap": "The event and the time reference overlap temporally"
"Before-Overlap": "The event began before the time reference and continues into it"
"Contains": "The time period contains the event entirely"
annotation_instructions: |
You will be shown a clinical note excerpt. Your task is to:
1. Highlight all time expressions (dates, times, durations) and event mentions
(procedures, conditions, medications, symptoms).
2. Classify the temporal relation between the primary event and time reference.
html_layout: |
<div style="padding: 15px; max-width: 800px; margin: auto;">
<div style="background: #f0f9ff; border: 1px solid #bae6fd; border-radius: 8px; padding: 16px; margin-bottom: 16px;">
<strong style="color: #0369a1;">Clinical Note:</strong>
<p style="font-size: 16px; line-height: 1.7; margin: 8px 0 0 0;">{{text}}</p>
</div>
</div>
allow_all_users: true
instances_per_annotator: 50
annotation_per_instance: 2
allow_skip: true
skip_reason_required: false
Sample Datasample-data.json
[
{
"id": "clintemp_001",
"text": "The patient underwent a cholecystectomy on March 15, 2016. Post-operative recovery was uneventful. Follow-up appointment scheduled for April 1."
},
{
"id": "clintemp_002",
"text": "Patient has been taking metformin 500mg twice daily since January 2015. Blood glucose levels have improved significantly over the past 6 months."
}
]
// ... and 8 more itemsGet This Design
Clone or download from the repository
Quick start:
git clone https://github.com/davidjurgens/potato-showcase.git cd potato-showcase/semeval/2016/task12-clinical-tempeval potato start config.yaml
Details
Annotation Types
Domain
Use Cases
Tags
Found an issue or want to improve this design?
Open an IssueRelated Designs
TimeLine: Cross-Document Event Ordering
Identify event mentions and temporal expressions in news text and classify temporal relations between events, based on SemEval-2015 Task 4 (Minard et al.). Annotators build timelines by ordering events related to a target entity across documents.
Analysis of Clinical Text: Disorder Identification and Normalization
Identify disorder mentions and their attributes in clinical discharge summaries, based on SemEval-2015 Task 14 (Elhadad et al.). Annotators mark disorder spans, body locations, severity indicators, and classify the assertion status of each disorder.
Aspect-Based Sentiment Analysis
Identification of aspect terms in review text with sentiment polarity classification for each aspect. Based on SemEval-2016 Task 5 (ABSA).