Κατεύθυνση Κλινική Φαρμακευτική
Library of the School of Science

2024-04-25

2024

Theofili Maria-Ileana

Καραλής Ευάγγελος, Αναπληρωτής Καθηγητής, Τμήμα Φαρμακευτικής ΕΚΠΑ

Εφαρμογή μεθόδων μηχανικής μάθησης για τη μελέτη των δερματικών αντιδράσεων υπερευαισθησίας σε φάρμακα και έκδοχα

Greek

Application of Machine Learning in skin hypersensitivity reactions to drugs and excipients

Skin is the largest organ of the human body and a major line of defense, fending off physical and chemical attacks. Skin reactions can be caused by various etiological factors, such as allergens or irritants and manifest with a wide range of symptoms like itching, redness and inflammation. The diagnosis of skin reactions is still a challenge and relies on three main pillars, a detailed medical history, physical examination and diagnostic tests. The main diagnostic method applied to manifestations of dermatitis is the patch test in which allergens are applied to areas of the skin where they remain for 2, 4 to 7 days and then the skin reactions are examined.
In the current study, patch test results from 240 patients with Allergic Contact Dermatitis were collected at the National Reference Center for Occupational Dermatoses "Andreas Syggros" Hospital (200 patients with positive reactions to ethylenediamine and 40 to budesonide). The MOAHLFA index was used to evaluate the data and descriptive statistics and statistical comparisons between groups were performed to study the results. The machine learning methods, Multiple Correspondence Analysis (MCA) and Principal Components Analysis for Categorical Data (CATPCA) were utilized to show the results and the entire statistical analysis was implemented in IBM SPSS® v.28.
Through the chi-square method were presented statistically significant correlations between the history of atopic dermatitis, professional activity and anatomical site with the gender of the volunteers. Statistically significant correlations were also found between hand and leg dermatitis with the occupational environment as well as between face and foot dermatitis with gender and age. The strong correlation between the values of the MOAHLFA index with the occupational class and another one with the anatomical site and the work environment was found due to MCA method.
Finally, the results of CATPCA demonstrated an association of increased age, especially in men, with the highest values of the MOAHLFA index and respectively of women and individuals of younger ages with low values of the dermatitis evaluation index.

Science

Health Sciences

Contact dermatitis, CD, Allergic dermatitis, ACD, Irritant dermatitis, ICD, allergens, irritants, Machine Learning, ML, MCA, CATPCA, Patch test, descriptive statistics, statistical comparisons, facial dermatitis, hand dermatitis, trunk dermatitis, foot dermatitis

No

0

Yes

136

139

File access is restricted until 2024-10-25.

https://pergamos.lib.uoa.gr/uoa/dl/object/3397374