Μοντέλα πρόβλεψης καρκίνου με χρήση γενετικών δεδομένων και τεχνικές μηχανικής μάθησης
Date Issued
February 10, 2020
Type
Μεταπτυχιακή Διπλωματική Εργασία
Abstract
Cancer is a disease with very high economic and social costs, without discrimination on the basis of age, sex, race or geographical origin. In particular, ovarian cancer is one of its insidious types because it has no symptoms until it reaches an advanced stage. This work discusses the identification of biometric markers that could be used to early diagnose or even prevent ovarian cancer using machine learning techniques. Clustering and classification algorithms are applied to data derived from mass / charge spectrometry measurements in order to identify proteomic patterns in blood serum that distinguish patients with ovarian cancer from those without and formulate rules that will accurately identify prospective patients. The purpose of this work is to compare classification algorithms in order to find the one that produces the best results. The results are also compared with other papers dealing with the same subject.
Subjects