Acne Severity Endpoints for Clinical Trials

The AI scoring provided by Legit.Health delivers automated, clinically validated acne severity scoring for clinical trials, quantifying inflammatory lesion count, spatial density, and IGA-aligned severity: the endpoints recommended by FDA guidance for establishing acne treatment effectiveness.

Acne Lesion And Density INdex

Score: 3.5

Report Information

☐

Timestamp

3/24/2026, 2:16:00 AM

☐

Analysis performed in

1.2 seconds

☐

Status

Not reviewed

☐

Body site

Left diagonal

☐

p]:mb-0>

Image quality

92%

☐

p]:mb-0>

Lesion count

36

☐

p]:mb-0>

Density

0.55

☐

p]:mb-0>

Local score

3.42

☐

Body site

Right diagonal

☐

p]:mb-0>

Image quality

88%

☐

p]:mb-0>

Lesion count

35

☐

p]:mb-0>

Density

0.6

☐

p]:mb-0>

Local score

3.49

FDA guidance context

The FDA’s guidance document Acne Vulgaris: Establishing Effectiveness of Drugs Intended for Treatment. Guidance for Industry. (2005) establishes the standard for clinical trial endpoints:

• Assessment of treatment effect should be based on both changes in lesion counts and success on the IGA, as these provide both quantitative and qualitative assessments of acne and useful complementary information.
• Inflammatory and non-inflammatory lesions should be counted and reported separately.
• The IGA should use a validated 5-point scale (0–4: Clear to Severe).
• Efficacy assessment should be limited to the face, as it is the most frequent site of involvement.

Endpoint capabilities

The AI provides four distinct endpoints, each configurable per study protocol:

Endpoint	Definition	AI output	Typical use in protocol
Inflammatory lesion count	Number of papules, pustules, and nodules detected	Integer count per perspective and total	Co-primary
Spatial density	Ratio of overlapping lesion detection areas to total detection area	Density score 0–1 per perspective	Exploratory
IGA score	Investigator Global Assessment severity	0–4 scale (Clear to Severe)	Co-primary
ALADIN composite	Combined lesion count and density metric (IGA × 2.5)	0–10 continuous scale	Exploratory

Why four endpoints?

The inflammatory lesion count and IGA are the two co-primary endpoints recommended by FDA guidance. Spatial density adds a dimension that lesion count alone misses; two patients with 30 lesions can have very different clinical presentations depending on whether those lesions are scattered or clustered in a concentrated area. The ALADIN composite (Acne Lesion And Density INdex) combines count and density into a single higher-resolution continuous scale, useful for detecting subtle treatment effects.

How the AI works

The scoring pipeline processes each uploaded photograph in three stages:

Stage 1: Inflammatory lesion detection

A deep learning object detection model identifies individual inflammatory acne lesions and draws bounding boxes around each one. The model detects papules, pustules, and nodules; comedones (non-inflammatory lesions) are excluded, consistent with IGA methodology and the FDA recommendation to count inflammatory and non-inflammatory lesions separately.

Lesion detection: bounding boxes identify each inflammatory acne lesion. The total count N is the number of detections.

Density calculation: overlapping detection circles indicate spatial concentration. Higher overlap = higher density.

Stage 2: Spatial density computation

The density score $D$ captures not just how many lesions are present, but how closely they cluster together:

$$D = \frac{A_{\text{overlap}}}{A_{\text{total}}}$$

where $A_{\text{overlap}}$ is the area of overlap between circular regions centered on detected lesions and $A_{\text{total}}$ is the total area covered by all detection regions. This distinction is clinically relevant: scattered mild acne and concentrated moderate acne can have identical lesion counts but very different severity perceptions.

Stage 3: IGA-aligned severity scoring

The lesion count and density are combined into a severity score aligned with the 5-point IGA scale:

$$\text{IGA} = N^a \cdot (D + b)$$

where $N$ is the lesion count, $D$ is the spatial density, and $a$, $b$ are empirically derived constants calibrated against the consensus of three board-certified dermatologists. The ALADIN composite score is then computed as IGA × 2.5, mapping to a 0–10 continuous scale.

IGA score	Severity	Clinical description
0	Clear	No inflammatory lesions
1	Almost clear	Rare inflammatory lesions with very low density
2	Mild	Some inflammatory lesions, low to moderate density, no nodules
3	Moderate	Many inflammatory lesions, moderate to high density, occasional nodules
4	Severe	Numerous inflammatory lesions, high density, many nodules

Visual report

The AI produces a structured severity report for each assessment:

Acne Lesion And Density INdex

Score: 3.5

Report Information

☐

Timestamp

3/24/2026, 2:16:00 AM

☐

Analysis performed in

1.2 seconds

☐

Status

Not reviewed

☐

Body site

Left diagonal

☐

p]:mb-0>

Image quality

92%

☐

p]:mb-0>

Lesion count

36

☐

p]:mb-0>

Density

0.55

☐

p]:mb-0>

Local score

3.42

☐

Body site

Right diagonal

☐

p]:mb-0>

Image quality

88%

☐

p]:mb-0>

Lesion count

35

☐

p]:mb-0>

Density

0.6

☐

p]:mb-0>

Local score

3.49

Protocol flexibility

The imaging and scoring protocol adapts to your study's needs. Legit.Health works with sponsors and CROs to configure the optimal protocol during study setup.

Imaging perspectives

Protocol	Perspectives	Use case
Standard (Hayashi Criterion)	2 views: Left diagonal (~70°), Right diagonal (~70°)	Most acne studies; captures majority of facial acne area per the Hayashi Criterion for counting inflammatory lesions per half-face.
Three-perspective	3 views: Left perpendicular, Frontal, Right perpendicular	Studies requiring full-face frontal coverage. The frontal view overlaps with both lateral views; facial landmark detection deduplicates lesions.
Custom	Any combination of perspectives	Any combination of perspectives, defined in collaboration with the sponsor during protocol design.

Global score aggregation

The system computes a local score (lesion count, density, IGA) for each perspective. The global score is derived from local scores using a configurable aggregation method:

Method	Formula	When to use
Maximum (default)	$\text{Global} = \max(\text{Local}_1, \text{Local}_2, \ldots)$	Captures the worst-affected area. Recommended for most acne protocols. When dermatologists perform a global IGA assessment, they are primarily influenced by the most severely affected region.
Sum	$\text{Global} = \sum \text{Local}_i$	Captures cumulative severity across the face. Useful when total burden matters.
Mean	$\text{Global} = \frac{1}{n}\sum \text{Local}_i$	Average severity. Useful when perspectives have significant overlap.

Overlap handling

When perspectives overlap, the AI uses Facial landmark detection to prevent double-counting:

1. Landmark identification: The AI detects facial landmarks (eyes, nose, mouth, jawline, forehead boundaries) in each image.
2. Region mapping: Each perspective is mapped to the facial regions it covers, based on the detected landmarks and the known capture angle.
3. Overlap detection: Lesions that appear in overlapping regions between two perspectives are identified using their spatial position relative to the facial landmarks.
4. Deduplication: Overlapping lesions are counted only once, attributed to the perspective with the highest confidence detection.

This enables flexible multi-perspective protocols without sacrificing accuracy.

Further reading

Scoring Methodology

Deep technical detail on the AI pipeline: lesion detection, density computation, IGA formula, multi-perspective scoring, and comparison with manual counting.

Imaging Protocol

Image capture protocols, DIQA quality control, perspective flexibility, facial landmark detection, and site standardisation.

Clinical Evidence

Peer-reviewed publications, validation metrics, performance claims, and the regulatory-grade evidence pathway.

Trial Workflow

End-to-end clinical trial workflow: protocol design, patient enrollment, image capture, scoring, data export, and EDC integration.

Publications

Peer-reviewed papers, conference presentations, key references from the literature, and the full clinical validation portfolio.