Národní úložiště šedé literatury Nalezeno 1 záznamů.  Hledání trvalo 0.01 vteřin. 
Automatic AIF voxel detection in DCE-MRI using machine learning
Frolíková, Štěpánka ; Jiřík, Radovan (oponent) ; Vitouš, Jiří (vedoucí práce)
Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) is a great tool for evaluating tissue perfusion. It requires accurate identification and detection of Arterial Input Function (AIF). Manual identification of AIF is not the fastest and the most effective method. Usually, automatic detection algorithms are not used. This thesis aims to precisely determine AIF using the trained machine learning model and compare the results with different approaches, like clustering. Data used for training and testing the model are both real and synthetic. The synthetic data are simulated using the DCATH pharmacokinetic model. The clustering method uses the K-means algorithm, optimized for human and mouse MRI images. The machine learning model uses a classifier based on the random forest method combined with clustering. The results evaluate this method’s accuracy and explain a model’s advantages and disadvantages. A functional and reliable automatic model will help to speed up the perfusion analysis and improve the quality of diagnosis.

Chcete být upozorněni, pokud se objeví nové záznamy odpovídající tomuto dotazu?
Přihlásit se k odběru RSS.