Sample Outputs and Deliverables

This page shows the concrete outputs that sponsors, CROs, and investigators receive from each alopecia severity assessment. The AI generates standardised SALT reports immediately after image submission.

Per-visit SALT report

Each set of four scalp images generates a complete SALT report with per-quadrant breakdowns, available within seconds of upload:

What the report contains

Output	Detail
Total SALT score	Global SALT score (0–100) with severity classification
Per-quadrant hair loss %	Left (18%), Right (18%), Top (40%), Back (24%)
Per-quadrant SALT contribution	Weighted SALT component per quadrant
Severity classification	None / Limited / Moderate / Severe / Very Severe
SALT response flags	SALT 50 / 75 / 90 / 100 vs. baseline (when applicable)
Image quality (DIQA)	Quality score per perspective; failed images flagged for recapture
Timestamp	UTC timestamp of capture and AI processing

SALT response flags for adverse event monitoring

The system automatically computes SALT responder status at every visit relative to baseline. For trials monitoring alopecia as a safety signal (e.g., MASH/MASLD trials), this provides objective, visit-by-visit documentation of any alopecia development without requiring manual SALT estimation.

Longitudinal SALT tracking and automated alerts

The platform tracks SALT score evolution across all visits and can trigger automated notifications when clinically significant changes are detected.

Longitudinal tracking

At each visit the system computes:

Absolute SALT change from baseline
Percentage change from baseline
SALT response classification: SALT 50 / 75 / 90 / 100 flag at each visit
Per-quadrant trend: Which scalp region is driving change
Score trajectory: Visual severity evolution across all timepoints

Automated alert system

The platform supports configurable automated alerts for safety monitoring:

Setting	Detail
Trigger	SALT increase ≥ 25% from baseline (configurable)
Recipient	Site email address
Threshold type	Inclusive (≥ 25%, not > 25%) to avoid missing borderline cases
Response	Investigator confirms finding clinically; escalates per protocol if confirmed

This alert system is particularly valuable for Phase 3 trials where alopecia is monitored as a treatment-emergent adverse event — sites are notified automatically without requiring manual data review.

MASH/MASLD safety deployment example

In a Phase 3 MASH/MASLD trial, investigators captured 4-perspective scalp images at scheduled clinic visits. SALT scores were computed automatically and monitored for any alopecia development related to the investigational therapy. The alert system flagged cases meeting the threshold for investigator review.

Data export for EDC integration

All assessment outputs are structured for export to the sponsor's EDC system. Fields exported at each visit:

Field	Description	Format
Patient identifier	Study-specific pseudonymised ID	String
Visit date	Date and timestamp of the assessment	ISO 8601
Total ASALT score	Global SALT score (0–100)	Float
Regional ASALT scores	ASALT score per quadrant (left, right, top, back)	Float (0–100) per region
Severity classification	None / Limited / Moderate / Severe / Very Severe	String
SALT response flags	Binary flags for SALT 50, SALT 75, SALT 90, SALT 100 response vs. baseline	Boolean per threshold
Change from baseline	Absolute and percentage change in ASALT from baseline	Float
DIQA scores	Image quality score per perspective	Float
Alert triggered	Whether the assessment triggered an automated notification based on threshold	Boolean

Data is transferred automatically via RESTful API or CSV/Excel export. Legit.Health provides IQ/OQ documentation and data mapping specifications for all major EDC platforms.

AI visual outputs

Beyond scores, the AI produces visual artefacts that demonstrate exactly what the system detected and how it reached its conclusion.

Hair loss segmentation masks

For each scalp quadrant, the AI generates a pixel-level segmentation mask classifying each region as hair-bearing or non-hair-bearing. This replaces visual estimation of hair loss percentage with an objective, continuous measurement.

Original photograph

Input: scalp quadrant photograph

Segmentation mask

Output: hair-bearing (green) vs. hair loss (red) regions

Overlay on original

Combined: segmentation overlaid on photograph

Hair follicle detection (trichoscopy)

When dermoscopic/trichoscopy images are available, the AI performs individual hair follicle detection with bounding boxes, providing follicle density measurements at sub-quadrant granularity.

Trichoscopy image

Input: dermoscopic scalp image

Follicle detection overlay

Output: individual follicles detected (mAP@50 = 0.816)

T0 → T1 evolution comparison

The platform enables side-by-side visual comparison of scalp quadrants across visits, with segmentation masks overlaid, so investigators can see exactly where hair loss has progressed or regressed.

Visit 1 (T0) segmentation

Baseline: SALT 55, segmentation mask

Visit 2 (T1) segmentation

Follow-up: SALT change from baseline

Reliability of the scores

The automated SALT system is validated at non-inferiority to expert inter-rater variability for segmentation and severity scoring. The AI produces the identical score for the same image every time, across every site, without calibration drift.

For the full validation evidence: Clinical Evidence →

Per-visit SALT report​

What the report contains​

Longitudinal SALT tracking and automated alerts​

Longitudinal tracking​

Automated alert system​

MASH/MASLD safety deployment example​

Data export for EDC integration​

AI visual outputs​

Hair loss segmentation masks​

Hair follicle detection (trichoscopy)​

T0 → T1 evolution comparison​

Reliability of the scores​