Quantifying Alopecia Areata via Texture Analysis to Automate the SALT Score Computation
December 2017
in “
Journal of Investigative Dermatology Symposium Proceedings
”
TLDR Researchers created a fast, accurate computer program to measure hair loss in alopecia areata patients.
In 2018, researchers developed a computer imaging algorithm to automate the Severity of Alopecia Tool (SALT) score computation for alopecia areata. The algorithm, which used texture analysis, was validated on a dataset of 250 images from over 100 pediatric patients aged 2-21 years. It demonstrated comparable accuracy to clinical assessments with a mean square error of 0.31 and was significantly faster, analyzing images in seconds rather than the average 7 minutes taken by clinicians. The study indicated that the algorithm could standardize and speed up the quantification of hair loss in alopecia areata, although it may not be specific to hair loss caused by alopecia areata alone. The research was supported by a National Alopecia Areata Pilot Grant and the National Alopecia Areata Foundation.