January 27, 2020
1 min read
Save

Principal component analysis used to characterize skin disease patterns

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

Principal component analysis can be used to characterize distinctive skin disease patterns using location and lesional data on cutaneous lupus erythematosus skin lesions, according to a cross-sectional study published in Journal of the American Academy of Dermatology.

Researchers used principal component analysis (PCA) to test if it may have the potential to identify clinical patterns of dermatologic diseases and help with subgroup classification. Cutaneous lupus erythematosus (CLE), which has well-described clinical subtypes, was used to test PCA.

Three hundred three patients with CLE were recruited from outpatient dermatology clinics at the University of Texas Southwestern Medical Center and Parkland Hospital in Dallas. A PCA of Cutaneous Lupus Disease Activity and Severity Index (CLASI) activity and damage component scores using SPSS version 25 was conducted. Five factors (F1 to F5) of unobserved constructs formed by sets of observed, correlated variables, using the sum scores method, were extracted.

“While the analysis largely correlated with known subtypes of CLE, this can also be used as a starting point to propose classification criteria for specific CLE disease subtypes, such as [subacute] CLE, for clinical trials,” Smriti Prasad, BSA, of the department of dermatology at the University of Texas Southwestern Medical Center in Dallas, and colleagues wrote.

F1 represented lesions on the anterior neck, chest, abdomen, arms and back/buttocks with high CLASI scores. F2 represented lesions on the ears and face with higher CLASI damage scores. F3 showed the posterior neck, back, arms and legs lesions with high damage scores. F4 showed hands and feet lesions with disease activity and damage, and F5 had disease activity and damage in the scalp measured by recent alopecia.

The study found that PCA can be used to characterize distinctive skin disease patterns through location and lesional data on CLE skin lesions.

“Larger multicenter studies are planned to confirm the association of the factors described here and identify other clinical phenotypes,” the researchers wrote. “We also propose that PCA can be utilized in other skin diseases with undefined clinical subtypes to identify clinical patterns that will help providers with diagnosis.” – by Erin T. Welsh

Disclosures: Prasad reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.