Hidradenitis Suppurativa Severity Endpoints for Clinical Trials
The AI scoring provided by Legit.Health delivers automated, clinically validated severity assessment for hidradenitis suppurativa (HS) clinical trials, computing the AIHS4 (Automated IHS4) score aligned with the International Hidradenitis Suppurativa Severity Score System.
Why automated IHS4 for clinical trials?
The IHS4 (International Hidradenitis Suppurativa Severity Score) is a validated, count-based severity measure that quantifies lesion burden across anatomical regions:
Manual IHS4 scoring is subject to inter-rater variability in lesion identification and counting, particularly for distinguishing abscesses from nodules and identifying draining tunnels. AIHS4 automates this through deep learning lesion detection, providing objective, reproducible counts.
How the AI works
Lesion detection and classification
A deep learning object detection model identifies three lesion types:
| Lesion type | IHS4 weight | AI approach |
|---|---|---|
| Papules/nodules | ×1 | Object detection and classification |
| Abscesses | ×2 | Differentiated from nodules by visual features (fluctuance, erythema pattern) |
| Draining fistulae (tunnels) | ×4 | Tract and opening detection |
Input: affected anatomical region
Output: bounding boxes by lesion type (colour-coded)
Papules (×1), abscesses (×2), fistulae (×4)
Severity classification
| IHS4 score | Severity |
|---|---|
| 0–3 | Mild |
| 4–10 | Moderate |
| ≥ 11 | Severe |
Validated reliability
AIHS4 achieves ICC = 0.727 (95% CI: 0.66–0.79) in the M-27134-01 clinical trial, substantially exceeding the manual inter-rater ICC of 0.47. See Clinical Evidence for full details.