Acne Severity Endpoints for Clinical Trials
The AI scoring provided by Legit.Health delivers automated, standardised acne assessment for clinical trials. It counts the inflammatory lesions on standardised photographs of the face, an objective count available on its own, and measures how densely they cluster. From these two measurements it computes ALADIN (Acne Lesion And Density INdex), a severity score that Legit.Health developed and published: a simple formula combines how many lesions there are and how clustered they are into a continuous score on the scale of the Investigator Global Assessment (IGA), from 0 to 4, and ALADIN reports the same score on a finer scale from 0 to 10. The global judgement a dermatologist makes becomes a reproducible measurement.
The platform supports interventional and observational studies across all clinical trial phases, from Phase I to Phase IV, as well as real-world evidence (RWE) and post-marketing studies.
IGA
3
ModerateALADIN (Acne Lesion And Density Index)
7.8
ModerateScale 0 – 10 (continuous)
Timestamp
2025-01-15 10:30 UTC (example)
Analysis performed in
1.2 seconds
Body site
Left diagonal
Image quality
92%
Lesion count
36
Density
0.61
Local score
3.21
Body site
Right diagonal
Image quality
88%
Lesion count
35
Density
0.58
Local score
3.06
The evidence behind these scores is on Clinical evidence.
The example images on this page are synthetic dermatology imagery used for illustration; they are not patient records.
Why automated acne scoring for clinical trials?
The FDA guidance on acne drug development bases efficacy on two measures: lesion counts and success on an investigator's global assessment (FDA, 2018). Both are hard to do consistently by hand. Counting dozens of small lesions on a face takes time, and the IGA rests on broad descriptors such as "some" or "many" lesions, so two raters looking at the same face often reach different grades (Medela et al., 2026).
The AI automates every component of the assessment:
- Lesion detection: every inflammatory lesion located and marked on each photograph, then counted
- Spatial density: how closely the lesions cluster, from 0 when they are scattered to close to 1 when they are confluent
- IGA score and ALADIN: count and density combined into a continuous score from 0 to 4 on the IGA scale, and into ALADIN, the same score from 0 to 10
- DIQA (Dermatology Image Quality Assessment): image quality checked at capture, so substandard images are retaken at the visit rather than surfacing as missing data at database lock
What is ALADIN, and why a new score?
Acne has more than 25 grading systems, and none of them is both objective and quick to use (Medela et al., 2026). Trials therefore rely on the IGA, a global judgement that is fast but subjective, and on lesion counts, which are objective but slow and say nothing about how concentrated the lesions are.
ALADIN (Acne Lesion And Density INdex) is a severity score that Legit.Health developed to bridge the two. It measures two things a dermatologist looks at when grading acne:
- How many inflammatory lesions there are: the lesion count
- How clustered they are: the spatial density, which tells scattered acne from acne packed into one area
A simple, published formula combines the two into a continuous score from 0 to 4, fitted to reproduce the IGA that expert dermatologists give the same photographs. ALADIN is that score multiplied by 2.5, on a scale from 0 to 10, which shows smaller changes between visits. Because every lesion that enters the score is marked on the image, the score can be explained, unlike a black-box classifier. How ALADIN is used in a trial, as supportive data alongside the investigator's IGA, is set out in Limitations.
- Reproducible: the same image always produces the same count, density and score, at every site and every visit.
- Traceable: every lesion that was counted is marked on the photograph, so the investigator and the monitor can see exactly what the number is made of.
- Continuous: ALADIN keeps the intermediate values that a whole IGA grade rounds away, which shows smaller changes between visits. Agreement with the investigator's IGA in a prospective study is being measured; see Clinical evidence.
Endpoint capabilities
The AI reports the lesion count and the ALADIN score from the same two photographs, and a protocol can take either or both:
- Output
- Whole number
- Per photograph, each lesion marked on the image; available without ALADIN
- What is counted
- One class
- Inflammatory lesions, without telling papules, pustules and nodules apart
- Spatial density
- 0 to 1
- How closely the detected lesions cluster on each photograph
- Score
- 0 to 10, continuous
- Lesion count and density combined in one published formula
- IGA score
- 0 to 4, continuous
- The same score on the IGA scale (ALADIN divided by 2.5); supportive data alongside the investigator's IGA
- Granularity
- Per photograph
- A score for each half of the face and a global score for the visit
The FDA guidance asks for both lesion counts and IGA success, with inflammatory and noninflammatory lesions counted separately. The platform reports the inflammatory lesion count, a continuous score on the IGA scale and ALADIN. Noninflammatory lesions (comedones) are not counted; see Limitations.
How the AI works
On each photograph, a deep learning model finds every inflammatory lesion and marks it. The software counts the detections, draws a circle around each one and measures how much the circles overlap: the more the lesions cluster, the higher the density. The count and the density then go into the ALADIN formula.
Left half of the face
| Measure | Value |
|---|---|
| Lesions | 36 |
| Density | 0.61 |
| IGA score (0 to 4) | 3.21 |
| ALADIN (0 to 10) | 8.0 |
Switch between the layers: the photograph as it was taken, every lesion the AI detected, and the circles used to measure density, darker where two or more of them overlap. Illustrative values.
Longitudinal severity tracking
From the follow-up visits onward, the same report shows how severity evolves across the study:
- Score trajectory: the lesion count and ALADIN at every visit, from screening to end of study
- Change from baseline: the absolute and percentage change in the inflammatory lesion count
- IGA success: whether the patient is clear or almost clear with at least a two-grade improvement from baseline, the definition in the FDA guidance; see response definitions
- Protocol adherence: whether assessments are captured at the correct intervals
The platform provides automated severity scoring as decision support. Scores are interpreted by qualified healthcare professionals within the patient's overall clinical context; the platform does not replace clinical judgement or make autonomous diagnostic decisions.