magistrsko delo
Abstract
Natančna in učinkovita segmentacija malignih limfomov na slikah pozitronske
emisijske tomografije/računalniške tomografije s fluorodeoksiglukozo (FDG PET/CT)
bi pripomogla pri pospešitvi in avtomatizaciji procesa segmentacije. Naloga raziskuje
uporabo metode globokega učenja nnUNet pri segmentaciji slik limfoma.
V nalogi so uporabljene slike FDG PET/CT 202 pacientov. Na slikah je bilo
naučenih 9 različnih modelov z različnimi učnimi množicami in hitrostmi učenja.
Z naučenimi modeli smo lahko zadovoljivo segmentirali območja limfomov, večja
območja so modeli segmentirali najboljše medtem ko manjše lezije limfomov slabše.
Izstopal je model, ki je imel največjo učno množico 143 pacientov in 10 h učenja,
kar je bil drugi najdaljši čas učenja in je tudi primerljiv po vrednostih metrike Dice
z ostalimi modeli segmentiranja tumorjev, ki jih najdemo v literaturi.
Ugotovitve te raziskave prispevajo k potencialu umetne inteligence pri kliničnem
delu ter ponazarjajo potencial metod globokega učenja, kot je nnUNet. Ker
se področje analize medicinskih slik še naprej razvija, integracija naprednih tehnik
umetne inteligence obeta bolj personalizirane in učinkovite zdravstvene rešitve.
Keywords
medicinska fizika;medicinsko slikanje;limfomi;pozitronska emisijska tomografija;računalniška tomografija;globoko učenje;
Data
Language: |
Slovenian |
Year of publishing: |
2023 |
Typology: |
2.09 - Master's Thesis |
Organization: |
UL FMF - Faculty of Mathematics and Physics |
Publisher: |
[A. Bavec] |
UDC: |
616-073 |
COBISS: |
169893635
|
Views: |
29 |
Downloads: |
6 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
Segmentation of lymphomas in PET/CT images with artificial intelligence |
Secondary abstract: |
Accurate and efficient segmentation of malignant lymphomas on fluorodeoxyglucose
positron emission tomography/computed tomography (FDG PET/CT) images
would help to speed up and automate the segmentation process. This thesis explores
deep learning method nnUNet in the context of lymphoma image segmentation.
FDG PET/CT images of 202 patients were used in this thesis. The images were
used to train 9 different models with different training sets and the duration’s of
learning. We were able to satisfactorily segment areas of lymphomas with trained
models, where larger areas of lymphoma involvement were better segmented by the
models while smaller lesions were associated with worse segmentation results. The
model that stood out was the one with the largest training set of 143 patients and
the duration of training of 10 h, which was the second longest training time, being
also comparable in terms of Dice metric values to other tumour segmentation models
found in the literature.
The findings of this research highlight potential of AI in clinical work and illustrate
the promising features of deep learning methods such as nnUNet. As the
field of medical image analysis continues to evolve, the integration of advanced AI
techniques promises more personalised and efficient healthcare solutions. |
Secondary keywords: |
medical physics;medical imaging;lymphoma;positron emission tomography;computed tomography;deep learning;nnUNet; |
Type (COBISS): |
Master's thesis/paper |
Study programme: |
0 |
Embargo end date (OpenAIRE): |
1970-01-01 |
Thesis comment: |
Univ. v Ljubljani, Fak. za matematiko in fiziko, Oddelek za fiziko |
Pages: |
68 str. |
ID: |
20488329 |