Přístupnostní navigace
E-application
Search Search Close
Bachelor's Thesis
Author of thesis: Lukáš Balog
Acad. year: 2025/2026
Supervisor: Ing. et Ing. Michal Nohel
Reviewer: Ing. Tomáš Vičar, Ph.D.
This bachelor thesis addresses the automatic segmentation of spinal multiple myeloma lesions in dual-energy CT data using deep learning methods. The input data were obtained from the publicly available Spinal-Multiple-Myeloma-SEG dataset, which was created using a dual-layer CT system (Philips IQon Spectral CT). For the purpose of this study, volumetric scans were utilized in the form of virtual monoenergetic images at 40~keV. Four deep learning architectures were implemented and compared nnU-Net, a residual encoder network (ResEnc U-Net), MedNeXt, and a residual attention network (ResAttUNet) each evaluated in both anisotropic and isotropic spatial resolution configurations. All models were trained and assessed using 5-fold cross-validation. Performance was evaluated using the Dice Similarity Coefficient (DSC), IoU, precision, recall, and detection rate, both at the patient level and at the per-lesion level stratified by lesion size category.
deep learning, convolutional neural networks, segmentation, myeloma lesions, spine, CT, dual-energy computed tomography, virtual monoenergetic imaging, nnU-Net, residual network
Date of defence
17.06.2026
Result of the defence
Defended (thesis was successfully defended)
Grading
C
Process of defence
Student prezentoval výsledky své práce a komise byla seznámena s posudky. Doc. Gumulec položil otázku: Dají se Vaše výsledky porovnat s jinou prací? Jak byste dosáhl lepších výsledků? Ing. Jakubíček položil otázku: Jaká byla vstupní data? Proč jste použil data jen z CT 40 keV? Student obhájil bakalářskou práci s výhradami a odpověděl na otázky členů komise a oponenta.
Language of thesis
Slovak
Faculty
Fakulta elektrotechniky a komunikačních technologií
Department
Department of Biomedical Engineering
Study programme
Biomedical Technology and Bioinformatics (BPC-BTB)
Composition of Committee
Doc. MUDr. Jaromír Gumulec, Ph.D. (předseda) doc. Mgr. Ing. Karel Sedlář, Ph.D. (místopředseda) Ing. Jan Odstrčilík, Ph.D. (člen) Ing. Jiří Sekora, MBA (člen) Ing. Andrea Němcová, Ph.D. (člen) Ing. Roman Jakubíček, Ph.D. (člen)
Supervisor’s reportIng. et Ing. Michal Nohel
Grade proposed by supervisor: C
Reviewer’s reportIng. Tomáš Vičar, Ph.D.
Grade proposed by reviewer: C
Responsibility: Mgr. et Mgr. Hana Odstrčilová