A comparative study of deep learning models for automated liver and tumor segmentation in 2d contrast-enhanced MRI images

Basit Öğe Kaydı
dc.authorid0000-0001-9840-4211
dc.authorid0009-0007-4392-3162
dc.authorid0009-0001-0667-3906
dc.authorid0009-0005-9172-8754
dc.authorid0009-0006-7039-7164
dc.authorid0000-0002-8909-2102
dc.authorid0000-0002-2126-8757
dc.contributor.authorTokatlı, Nazlı
dc.contributor.authorBilmez, Yakuphan
dc.contributor.authorBayram, Mücahit
dc.contributor.authorBayır, Beyzanur
dc.contributor.authorÖzalkan, Helin
dc.contributor.authorTekin, Zeynep
dc.contributor.authorÖrmeci, Necati
dc.contributor.authorAltun, Halis
dc.date.accessioned2026-01-22T15:09:53Z
dc.date.available2026-01-22T15:09:53Z
dc.date.issued2025
dc.departmentFakülteler, Mühendislik ve Doğa Bilimleri Fakültesi, Bilgisayar Mühendisliği Bölümü
dc.departmentFakülteler, Mühendislik ve Doğa Bilimleri Fakültesi, Yazılım Mühendisliği Bölümü
dc.departmentFakülteler, Tıp Fakültesi, Dahili Tıp Bilimleri Bölümü, İç Hastalıkları Ana Bilim Dalı
dc.description.abstractThis paper presents a comprehensive investigation into deep learning techniques for the automated segmentation of the liver and tumors from 2D abdominal contrast-enhanced Magnetic Resonance Imaging (MRI) slices. Addressing a significant challenge in medical image analysis, our study leverages the public ATLAS dataset [1], using a selection of 60 3D abdominal MRI scans, from which we extracted approximately 3,750 2D slices for model training and evaluation. The core objective was the precise identification and delineation of both the liver organ and any intrahepatic lesions. A comparative analysis was conducted on three U-Net-based architectures: the standard Attention U-Net model incorporating EfficientNet-b3 and CBAM but without Focal Loss, the Attention U-Net model with integrated Focal Loss, and the ResNet34-Based U-Net model. To optimize performance, we explored the efficacy of different loss functions, namely DiceLoss and a hybrid DiceLoss with Focalcoss. Our findings are promising: Among the evaluated models, the ResNet34-Based U-Net demonstrated the highest performance with a Dice score of 91.36% and an IoU score of 89.52%. It was followed by the Attention U-Net with Focal Loss, which achieved 86.41% Dice and 81.61% IoU scores, and the standard Attention U-Net, which obtained 85.93% Dice and 81.19% IoU scores. These results underscore the significant potential of our 2D-based methodology to enhance the precision and efficiency of liver and tumor detection from abdominal scans, offering a valuable tool to support clinicians in early diagnosis and to alleviate their workload.
dc.identifier.citationTokatlı, N., Bilmez, Y., Bayram, M., Bayır, B., Özalkan, H., Tekin, Z., Örmeci, N., & Altun, H. (2025). A comparative study of deep learning models for automated liver and tumor segmentation in 2d contrast-enhanced MRI images. 2025 Innovations in Intelligent Systems and Applications Conference, ASYU 2025, (ss. 1-6). IEEE. https://doi.org/10.1109/ASYU67174.2025.11208455
dc.identifier.doi10.1109/ASYU67174.2025.11208455
dc.identifier.endpage6
dc.identifier.isbn9798331597276
dc.identifier.issn2770-7946
dc.identifier.issn2770-7938
dc.identifier.scopus2-s2.0-105022467095
dc.identifier.startpage1
dc.identifier.urihttps://doi.org/10.1109/ASYU67174.2025.11208455
dc.identifier.urihttps://hdl.handle.net/20.500.13055/1262
dc.indekslendigikaynakScopus
dc.institutionauthorTokatlı, Nazlı
dc.institutionauthorBilmez, Yakuphan
dc.institutionauthorBayram, Mücahit
dc.institutionauthorBayır, Beyzanur
dc.institutionauthorÖzalkan, Helin
dc.institutionauthorTekin, Zeynep
dc.institutionauthorÖrmeci, Necati
dc.institutionauthorAltun, Halis
dc.institutionauthorid0000-0001-9840-4211
dc.institutionauthorid0009-0007-4392-3162
dc.institutionauthorid0009-0001-0667-3906
dc.institutionauthorid0009-0005-9172-8754
dc.institutionauthorid0009-0006-7039-7164
dc.institutionauthorid0000-0002-8909-2102
dc.institutionauthorid0000-0002-2126-8757
dc.language.isoen
dc.publisherIEEE
dc.relation.ispartof2025 Innovations in Intelligent Systems and Applications Conference, ASYU 2025
dc.relation.publicationcategoryKonferans Öğesi - Uluslararası - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.subjectDeep Learning
dc.subjectLiver Tumor Segmentation
dc.subjectMagnetic Resonance Imaging
dc.subjectMedical Imaging
dc.subjectU-Net Architectures
dc.titleA comparative study of deep learning models for automated liver and tumor segmentation in 2d contrast-enhanced MRI images
dc.typeConference Object
dspace.entity.typePublication

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Kapalı Erişim
İsim:
Tam Metin / Full Text.pdf
Boyut:
653.82 KB
Biçim:
Adobe Portable Document Format
İndir
Lisans paketi
Listeleniyor 1 - 1 / 1
Kapalı Erişim
İsim:
license.txt
Boyut:
1.17 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

Bilgisayar Mühendisliği Bölümü Koleksiyonu
Dahili Tıp Bilimleri Bölümü Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu
Yazılım Mühendisliği Bölümü