Multivariate statistical analysis for dermatological disease diagnosis

Abstract

The differential diagnosis of some erythemato-squamous diseases is a major problem in dermatology. This is the case with: psoriasis, seborrhoeic dermatitis, lichen planus, pityriasis rosea, chronic dermatitis, and pityriasis rubra pilaris. Further complicating diagnosis, they all share clinical features, with very few differences. Although biopsies could help physicians, these diseases also share many histopathological features. In this context, this research applies a multivariate statistical analysis to explore the Dermatology Data Set (available in the UCI data repository) and construct a classifier, based on the clinical features, as an aid to the medical diagnosis of erythemato-squamous dermatological diseases. The research results provide enhanced knowledge that can help to enrich dermatological diagnoses made by doctors. Also, the classifier developed using the Linear Discriminant Analysis obtains a high mean accuracy rate in relation to the 6 diseases (83.73% correct classifications). This rate means that patients have a strong chance of being treated adequately, while biopsies may also be solicited.