Dissertation committee:
Αλέξανδρος Στρατηγός, Καθηγητής, Ιατρική Σχολή, Εθνικό και Καποδιστριακό Πανεπιστήμιο Αθηνών, Επιβλέπων
Δημήτριος Ρηγόπουλος, Καθηγητής, Ιατρική Σχολή, Εθνικό και Καποδιστριακό Πανεπιστήμιο Αθηνών
Ηλέκτρα Νικολαΐδου, Καθηγητής, Ιατρική Σχολή, Εθνικό και Καποδιστριακό Πανεπιστήμιο Αθηνών
Σταμάτιος Γρηγορίου, Καθηγητής, Ιατρική Σχολή, Εθνικό και Καποδιστριακό Πανεπιστήμιο Αθηνών
Αργυρώ Χατζηϊωάννου, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, Εθνικό και Καποδιστριακό Πανεπιστήμιο Αθηνών
Αιμίλιος Λάλλας, Αναπληρωτής Καθηγητής, Ιατρική Σχολή, Αριστοτέλειο Πανεπιστήμιο Θεσσαλονίκης
Ashfaq A. Marghoob, Assistant Professor, Memorial Sloan Kettering Cancer Center
Summary:
Artificial Intelligence (AI) has been incorporated in a wide spectrum of our daily lives, and Medicine could be no exemption to it. A plurality of scientific articles has explored the application of AI in fields such as Ophthalmology and Radiology, while there are already FDA approved, AI applications, which are used in clinical practice. Through our research we explored: (1) Can AI be used in early skin cancer diagnosis? (2) Which are the pitfalls of AI algorithms in Dermatology and in which possible ways could they be improved upon? (3) How could AI be of use in Dermatology, and in which ways could it be used? (4) Which is the best approach to the research conducted with regards to AI algorithms in early skin cancer diagnosis, and how should these results be interpreted? And (5), how can we improve the diagnostic accuracy of clinicians and AI algorithms for early skin cancer diagnosis and more specifically, melanoma? Finally, we attempted to describe the overall framework, within which, AI algorithms could be proven useful in clinical practice, and more importantly, beneficial to the patients.