Saltar al contenido principal

Publications and References

Primary validation study

"Evaluation of AIHS4 Performance in the M-27134-01 Clinical Trial for Hidradenitis Suppurativa" (2023)

Observational non-interventional study evaluating the performance and reliability of the AIHS4 scoring system within the context of a Phase 2/3 clinical trial for hidradenitis suppurativa. The study demonstrated ICC = 0.727, substantially exceeding the manual inter-rater benchmark of ICC 0.47.

Image quality validation

"Dermatology Image Quality Assessment (DIQA): Artificial intelligence to ensure the clinical utility of images for remote consultations and clinical trials" Hernández Montilla, I., Mac Carthy, T., Aguilar, A., Medela, A. Journal of the American Academy of Dermatology, Volume 88, Issue 4, pp. 927–928 (2023) DOI: 10.1016/j.jaad.2022.11.002 | PMID: 36526082

DIQA is the image quality assessment algorithm that acts as a quality gate in the clinical trial workflow.

The same AI architecture used for HS scoring has been validated across multiple conditions:

StudyConditionEndpointKey metricStatus
ALADIN (Sabater et al., 2026)AcneIGA, lesion countCohen's κ = 0.53Published
APASIPsoriasisPASI componentsRMAE ≤ 0.153Published
Automated SALTAlopeciaSALT scoringRMAE = 7.08%Deployed in Phase 3
ASCORAD (Medela et al., 2022)Atopic dermatitisSCORAD scoringPilot validatedPublished
aEASI_HVNAtopic dermatitisEASI scoringOngoing

Key references from the literature

IHS4 methodology

  • Zouboulis, C. C. et al. Development and validation of the International Hidradenitis Suppurativa Severity Score System (IHS4), a novel dynamic severity assessment tool. British Journal of Dermatology, 177(5), 1401–1409 (2017). — The original IHS4 development and validation paper.

Hurley staging

  • Hurley, H. J. Axillary hyperhidrosis, apocrine bromhidrosis, hidradenitis suppurativa, and familial benign pemphigus: surgical approach. In: Roenigk, R. K., Roenigk, H. H., eds. Dermatologic Surgery. Marcel Dekker, 729–739 (1989).

Inter-rater reliability in HS assessment

  • Literature reports IHS4 inter-observer ICC of approximately 0.47 (95% CI: 0.32–0.65), reflecting the inherent difficulty of distinguishing lesion types (particularly abscess vs. nodule) across independent raters. This variability is the primary motivation for automated scoring.