University of Patras
Κλινικά και δερματοσκοπικά χαρακτηριστικά μελανοκυτταρικών αλλοιώσεων του δέρματος: συσχέτιση με τα ιστοπαθολογικά ευρήματα και ανάλυση εικόνας
Abstract
dc:descriptionBackground: An accurate clinical classification of common benign nevi, atypical nevi and melanomas is of outstanding importance. The non-invasive method of dermoscopy, allowed for the visualization of structures invisible to the naked eye, and undoubtedly advanced the assessment of melanocytic lesions to a new dimension. Machine learning, asubfield of artificial intelligence, finds application in dermatology using imaging data to build models through training and testing processes. Aim of this study was to correlate the sensitivity and specificity of naked-eye examination and dermoscopy, in diagnosing melanocytic lesions, compared to the histopathological results, constituting the gold standard of diagnosis, but also to evaluate the reliability of machine learning in classifying melanocytic lesions. Material and Methods: One hundred eighteen melanocytic lesions were clinically evaluated via the naked eye and dermoscopic examination, using Pattern Analysis Methodology, and afterward, the clinical and dermoscopic photographs were taken and stored. Subsequently, the lesions were excised and histologically examined. Findings from the clinical and dermoscopic assessement were correlated with the histological findings. Then, 127 feautures were extracted from clinical images for all three categories of melanocytic lesions; high risk, medium risk and low risk, to assess the existence of a possible statistical significant difference between the three categories for each feature. Logistic regression classifier, random forest classifier, and support vector classifier were used to evaluate the diagnostic performance of each feature. The performance of each classifier was evaluated through ROC analysis. Results: According to the final histopathological analysis, 63 common benign nevi, 41 dysplastic nevi, and 14 cutaneous melanomas were excised in total. Clinical examination via the naked eye showed 78.2% sensitivity and 71.4% specificity in identifying the clinical atypia, while dermoscopy demonstrated 89.1% sensitivity and 93.7% specificity. The non-parametric Kruskal-Wallis analysis revealed 95 statistically significant characteristics (p-values<0.05). Of these 95 statistically significant feautures, post-hoc analysis showed that in 66 cases the differentiation is statistically significant (p-value < 0.05), for the class pair (LR-MR), in 56 cases for the class pair (HR-MR), whereas in 50 cases the differentiation is statistiacally significant (p-value <0.05) for the class pair (LR-HR). The logistic regression classifier has the best performance across all possible class pairs, while the support vector machine classifier has the worst performance over the other two. Conclusions: The results of the present study indicate a higher sensitivity and specificity of dermoscopy in evaluating and diagnosing melanocytic lesions, compared to the naked-eye examination. Machine learning could be used as an additional tool, when evaluating melanocytic lesions.
Degree
thesis:*- Grantor dc:publisher
- University of Patras
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kalloniati, Evangelia
- Καλλονιάτη, Ευαγγελία
Subjects
dc:subject × 14Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/55868