| 摘要 |
<p>The invention relates to an intelligent clinical decision support system for breast cancer diagnosis by non-invasive methods. According to the invention, the system comprises a sub-system (A) for generating a mathematical model and an exploitation sub-system (B); the first sub-system (A) comprises, in its turn, a block (1) for inserting data i, k=1,...,4 and j=1,...,79, consisting of the value (i) of the KGF molecule concentration for the patient j, the value (i) of the Angiopoi2 molecule concentration for the patient j, the value (i) of the PDGF-bb molecule concentration for the patient j and the age (i) for the patient j, a unit (2) for random dividing the input data set into training data (a), test data (t) and validation data (v), a unit (3) for processing with genetic algorythms, which offers a selected program which is validated in a block (6), and after validation the program offers a mathematical model (5) which is described in a programming language, thereby obtaining a program which is used for the diagnosis of breast cancer, within the second exploitation sub-system (B) which comprises a calculus module which implements the mathematical model (5), the obtained results being displayed on a display module (10).</p> |
