Search

Your search keyword '"Gumus, Kazim Z"' showing total 10 results
Start Over Author "Gumus, Kazim Z"
10 results on '"Gumus, Kazim Z"'

5. Deep learning algorithms for bladder cancer segmentation on multi-parametric MRI

Author

Gumus, Kazim Z., Nicolas, Julien, Gopireddy, Dheeraj R., Dolz, Jose, Jazayeri, Seyed Behzad, Bandyk, Mark, Gumus, Kazim Z., Nicolas, Julien, Gopireddy, Dheeraj R., Dolz, Jose, Jazayeri, Seyed Behzad, and Bandyk, Mark

Abstract

Background: Bladder cancer (BC) segmentation on MRI images is the first step to determining the presence of muscular invasion. This study aimed to assess the tumor segmentation performance of three deep learning (DL) models on multi-parametric MRI (mp-MRI) images. Methods: We studied 53 patients with bladder cancer. Bladder tumors were segmented on each slice of T2-weighted (T2WI), diffusion-weighted imaging/apparent diffusion coefficient (DWI/ADC), and T1-weighted contrast-enhanced (T1WI) images acquired at a 3Tesla MRI scanner. We trained Unet, MAnet, and PSPnet using three loss functions: cross-entropy (CE), dice similarity coefficient loss (DSC), and focal loss (FL). We evaluated the model performances using DSC, Hausdorff distance (HD), and expected calibration error (ECE). Results: The MAnet algorithm with the CE+DSC loss function gave the highest DSC values on the ADC, T2WI, and T1WI images. PSPnet with CE+DSC obtained the smallest HDs on the ADC, T2WI, and T1WI images. The segmentation accuracy overall was better on the ADC and T1WI than on the T2WI. The ECEs were the smallest for PSPnet with FL on the ADC images, while they were the smallest for MAnet with CE+DSC on the T2WI and T1WI. Conclusions: Compared to Unet, MAnet and PSPnet with a hybrid CE+DSC loss function displayed better performances in BC segmentation depending on the choice of the evaluation metric.

Published
2024

6. MRI‐based radiomic features for identifying recurrent prostate cancer after proton radiation therapy

Author

Gumus, Kazim Z., primary, Contreras, Samuel Serrano, additional, Al‐Toubat, Mohammed, additional, Harmon, Ira, additional, Hernandez, Mauricio, additional, Ozdemir, Savas, additional, Kumar, Sindhu, additional, Yuruk, Nurcan, additional, Mete, Mutlu, additional, Balaji, K. C., additional, Bandyk, Mark, additional, and Gopireddy, Dheeraj R., additional

Published
2024
Full Text
View/download PDF

10. Search for new physics in the compact muon solenoid (CMS) experiment and the response of the CMS calorimeters to particles and jets

Author

Gumus, Kazim Z, Borst, Walter L., Lodhi, M. A. K., Wigmans, Richard, and Akchurin, Nural

Subjects
Jet, Large hadron collider (LHC), Compact muon solenoid (CMS), Dijet, Resonance, Standard model
Abstract

A Monte Carlo study of a generic search for new resonances beyond the Standard Model (SM) in the CMS experiment is presented. The resonances are axigluon, coloron, E6 diquark, excited quark, W', Z', and the Randall-Sundrum graviton which decay to dijets. The dijet resonance cross section that the CMS can expect to discover at a 5 sigma significance or to exclude at 95% confidence level for integrated luminosities of 100 pb−1, 1 fb−1, and 10 fb−1 is evaluated. It is shown that a 5 sigma discovery of a multi-TeV dijet resonance is possible for an axigluon, excited quark, and E6 diquark. However, a 5 sigma discovery can not be projected with confidence for a W', Z' and the Randall-Sundrum graviton. On the other hand, 95% CL exclusion mass regions can be measured for all resonances at high luminosities. In the second part of this dissertation, the analyses of the 2006 test beam data from the combined electromagnetic and hadronic barrel calorimeters are presented. The CMS barrel calorimeters’ response to a variety of beam particles in a wide momenta range (1 to 350 GeV/c) is measured. Furthermore, using these beam data, the expected performance of the barrel calorimeters to jets is predicted.

Published
2008

Catalog

Books, media, physical & digital resources
See catalog results