Κατεύθυνση Κλινική Φαρμακευτική
Library of the School of Science

2024-05-16

2024

Diavoli Spyridoula

Ε. Καραλής, Αναπλ. Καθηγητής Τμήμα Φαρμακευτής ΕΚΠΑ (Επιβλέπων),
Μ. Βλάχου, Αναπλ. Καθηγήτρια Τμήμα Φαρμακευτής ΕΚΠΑ,
Σ. Μαρκαντώνη - Κυρούδη, Καθηγήτρια Τμήμα Φαρμακευτής ΕΚΠΑ

« ΕΦΑΡΜΟΓΗ ΜΕΘΟΔΩΝ ΤΕΧΝΗΤΗΣ ΝΟΗΜΟΣΥΝΗΣ ΣΕ ΦΑΡΜΑΚΟΚΙΝΗΤΙΚΑ ΔΕΔΟΜΕΝΑ »

Greek

APPLICATION OF ARTIFICIAL INTELLIGENCE METHODS IN PHARMACOKINETIC DATA

This thesis deals with the use of artificial intelligence for the analysis of pharmacokinetic
data.More specifically, the use of artificial intelligence in pharmacokinetics has opened up new
perspectives for the analysis and interpretation of these data. The thesis focuses on the examination
of various artificial intelligence methods, such as machine learning algorithms, for the analysis
and prediction of pharmacokinetic parameters, with Random Forest and Principal Component
Analysis being the most important representatives.Specifically, the paper analyzes three datasets
through which the pharmacokinetic parameters necessary for this study, such as area under the
curve (AUC) and time of maximum concentration (Cmax), were calculated. Then two methods of
artificial intelligence data analysis are used: first Principal Component Analysis and then Random
Forest. This can be applied to a dataset containing information from previous bioequivalence
studies and then performed on new data.The results of the work demonstrate that the use of
artificial intelligence can lead to an understanding of the relationship between pharmacokinetic
parameters and what each actually expresses. This will lead to the development of parameters that
correctly express the extent and rate of absorption, which can then be incorporated for use in
bioequivalence studies and the personalization of pharmacotherapy.Finally, the work offers
prospects for future extensions and improvements in this area. Further research can focus on
examining different machine learning algorithms, incorporating more parameters and data, and
evaluating the effectiveness of the proposed methods in clinical studies.

Science

Machine learning, artificial intelligence, deep learning, Alan Turing

Yes

2

Yes

33

56

File access is restricted only to the intranet of UoA.

https://pergamos.lib.uoa.gr/uoa/dl/object/3398779