Recurrence risk factor extraction

Skill KangjieShen/openscar-starter/skills/research/recurrence-risk-factor-extraction

AI-native scar and keloid phenotyping framework with registry schema, synthetic mock data, validation scripts, and clinician-reviewable workflow skills

Install

npx -y skills add KangjieShen/openscar-starter --skill recurrence-risk-factor-extraction

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Extract candidate keloid recurrence risk factors from de-identified chart text for clinician-reviewed OpenScar registry entry.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.1 KB, as published. Nobody here has run it

Recurrence Risk Factor Extraction

Description

Use this skill to extract structured candidate recurrence risk factors from de-identified keloid or scar chart text. It prepares research variables for clinician review and registry entry. It does not predict recurrence or assign risk scores.

When to Use

A retrospective note needs variables for recurrence or treatment-response analysis.
A case series needs consistent abstraction of lesion, patient, treatment, and follow-up factors.
A reviewer needs missing-variable questions before data entry.

Do Not Use

Do not generate patient-specific recurrence predictions.
Do not treat extracted factors as causal.
Do not use identifiable records in public outputs.
Do not infer sensitive demographics unless documented and necessary.

Required Inputs

De-identified chart text or structured notes.
Target lesion ID if multiple lesions are present.
Time window for recurrence assessment.
Local recurrence definition, if available.

Workflow

Confirm the note is de-identified.
Extract patient-level context:
- age at first visit
- sex at birth when documented
- Fitzpatrick skin type when documented
- prior keloid history
- family history
- wound-healing comorbidities
Extract lesion-level context:
- scar category
- anatomic site
- laterality
- presumed trigger
- duration at baseline
- baseline symptoms
Extract treatment context:
- modality
- combination treatment
- adverse events
- treatment timing relative to baseline
Extract outcome context:
- follow-up duration
- response category candidate
- recurrence status candidate
- recurrence definition
- confidence and missing fields
Produce a candidate variable table and a missing-data checklist.

Output Format

Extracted candidate risk factors
Source phrases or evidence snippets, paraphrased when needed
Suggested OpenScar table and field
Missing variables
Ambiguities for clinician review
Research-use warning

Safety and Compliance

This skill supports documentation, research preparation, and clinician review. It does not diagnose, prescribe, determine recurrence, or replace licensed clinical judgment. Escalate urgent symptoms, suspected infection, wound breakdown, severe pain, rapid progression, bleeding, ulceration, or functional compromise to a qualified clinician.

Use de-identified text and avoid exporting direct identifiers, full dates, addresses, contact details, or identifiable image links.

Examples

Extract recurrence-related variables from this de-identified keloid chart note for OpenScar registry review.

Integration

Works well with keloid-follow-up-intake, scar-treatment-response-review, scar-registry-entry-builder, and scar-literature-review.