Inside Point-and-Capture Wound Imaging: How AI Measurement Works

Manual wound measurement with a ruler and cotton swab is inconsistent between clinicians and slow to trend. Point-and-capture imaging replaces it with a phone camera and a computer vision model. Here's how it actually works.

Border detection

When you point the camera at the wound, a segmentation model identifies wound edges in real time. No calibration sticker required for typical anatomy — depth cues and reference geometry are inferred from the capture. See AI-powered wound imaging for supported wound types.

Measurement outputs

Each capture produces:

• Length and width along the wound's longest and perpendicular axes
• Surface area (cm²) from the detected border
• Depth estimation where the model can resolve it
• Tissue composition breakdown (granulation, slough, eschar, epithelial)

Why point-and-capture matters for trajectory

Consistent measurement is the foundation of healing trajectory analytics. When every capture uses the same model, week-over-week percent area reduction becomes reliable — which is what CMS looks for when justifying continued advanced therapy.

Integration with the ambient scribe

The capture doesn't sit alone. Measurements pass into the SOAP note automatically, so the assessment reflects real numbers rather than clinician memory. See the joint workflow in Point-n-Capture deep dive.

Accuracy and clinician oversight

All measurements are editable. The clinician remains the source of truth — the model handles the tedium. Common accuracy questions are answered in the AI wound measurement FAQ.

Field use

Captures work offline for mobile wound care providers and sync when signal returns — so route-based clinicians aren't blocked between homes or facilities.