ASALT
What is ASALT?
The Automatic Severity of Alopecia Tool (ASALT) is the automated implementation of the Severity of Alopecia Tool (SALT), one of the most extensively employed scoring systems for evaluating alopecia areata severity. The main constraint of this scoring system lies in the issue of inter-observer variability, as it is complex for a human observer to compute hair loss percentages.
Addressing this challenge, ASALT uses smartphone-captured images. The algorithm identifies the extent and percentage of hair loss, enabling the computation of the SALT score.
Body site
To compute the SALT, up to four photographs are needed, covering the following views of the head:
- Top (HEAD_TOP)
- Left side (HEAD_LEFT)
- Right side (HEAD_RIGHT)
- Back (HEAD_BACK)
Each site contributes a different percentage to the final score: top 40%, left and right side 18% each and the back 24%.
Local ASALT
As the ASALT works at image-level, it is possible to perform the follow-up of a single zone, for example, the top of the head. The output of the ASALT contains two main outputs, the percentage from 0 to 100 of the hair loss in that specific body site and also, the weighted percentage, which takes into account the body site.
General ASALT
The general ASALT is computed by adding multiple photos of the head into the equation. Only photos of affected areas are required, and the procedure for acquisition is explained in the next section.
How to take pictures
Image quality affects algorithm performance. A guide with essential tips for image acquisition is available. This section is dedicated to the photo capture process for ASALT computation.
How does ASALT work?
ASALT uses convolutional neural networks trained on a dataset comprising thousands of images annotated by experts. The algorithm quantifies the extent of low and high hair density and computes a percentage of hair loss.
Request ASALT score
{
"requestId": "90925097-820b-403d-a75d-4cd989903df1",
"data": {
"type": "image",
"modality": "clinical",
"operator": "Practitioner",
"bodySite": "HEAD_TOP",
"knownConditionForThisImage": {
"conclusion": "Alopecia areata"
},
"subject": {
"identifier": "6ec724a0-6fa3-11eb-a15f-0242ac160004",
"gender": "m",
"height": 175,
"weight": 71,
"birthdate": "1986-10-21",
"generalPractitioner": {
"identifier": "44f89a8c-6f8a-11eb-9c8a-0242ac160004"
},
"managingOrganization": {
"identifier": "b13cd636-327b-11ec-86b0-0242ac180004",
"display": "Hospital Central"
}
},
"scoringSystems": ["ASALT", "DLQI"],
"questionnaireResponse": {
"DLQI": {
"question1": 2,
"question2": 3,
"question3": 1,
"question4": 2,
"question5": 1,
"question6": 2,
"question7": 2,
"question8": 2,
"question9": 3,
"question10": 2
}
},
"content": "base64 image"
}
}
The bodySite should be specified as one of the following options:
- HEAD_TOP
- HEAD_LEFT
- HEAD_RIGHT
- HEAD_BACK
Response
"evolution": {
"domains": {
"ASALT": {
"explainabilityMedia": {
"content": "base 64 image",
"detections": null
},
"facets": {
"alopecia": {
"intensity": 100,
"value": 100
}
},
"grade": {
"category": null,
"score": 40
}
}
}
}
The ASALT does not have a translation into the more clinician-friendly concepts of “mild”, “moderate”, “severe” or “very severe”, and therefore a null output is given for the category key. More information about the lack of a broad agreement can be found in the Defining Severity in Alopecia Areata: Current Perspectives and a Multidimensional Framework study.