The potential of radiomics features in the detection of hepatocellular carcinoma (HCC) in 2D liver MRI images by using machine learning methods

Basit Öğe Kaydı
dc.contributor.authorTulu, Cagatay Neftali
dc.contributor.authorIbrikci, Turgay
dc.date.accessioned2026-02-27T07:32:50Z
dc.date.available2026-02-27T07:32:50Z
dc.date.issued2025
dc.description.abstractHepatocellular carcinoma (HCC) is the sixth most prevalent cancer globally and the fourth leading cause of cancer-related mortality. Early detection of tumors through non-invasive methods, such as MRI (Magnetic Resounance Imaging), is critical for diagnosis and treatment planning. This study proposes a machine learning-based approach to predict the presence of HCC tumors in two-dimensional liver MR images by employing radiomics features extracted from the images. The publicly available Atlas dataset was used to extract radiomics features from MRI slices taken from 60 patients, those features were then employed to train machine learning models. The results demonstrate that radiomics features are highly effective for detecting HCC tumors. The LightGBM machine learning classifier achieved an AUC score of 0.95 and an Accuracy score of 0.87, indicating strong predictive accuracy and efficiency in identifying tumor-containing images. Highlighting the Center of Mass Index and Gray Level Size Zone Matrix (GLSZM) features are highly effective radiomics features in the model decisions. The performance of the trained models confirms that extracting radiomics features from MRI images is an effective and reliable approach for automatically detecting HCC tumors. The obtained results indicate promising outcomes, suggesting that the diagnostic process using radiomics features can be successfully applied to medical images in various clinical fields.
dc.description.sponsorshipTUBITAK [1001-122S588]
dc.description.sponsorshipThis project study is supported by TUBITAK-1001-122S588.
dc.identifier.doi10.1007/s11760-025-04162-z
dc.identifier.issn1863-1703
dc.identifier.issn1863-1711
dc.identifier.issue7
dc.identifier.urihttp://dx.doi.org/10.1007/s11760-025-04162-z
dc.identifier.urihttps://hdl.handle.net/20.500.14669/4354
dc.identifier.volume19
dc.identifier.wosWOS:001486307900027
dc.indekslendigikaynakWeb of Science
dc.language.isoen
dc.publisherSpringer London Ltd
dc.relation.ispartofSignal Image and Video Processing
dc.relation.publicationcategoryMakale - Uluslararas� Hakemli Dergi - Kurum ��retim Eleman�
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.snmzKA_20260302
dc.subjectHCC tumor detection
dc.subjectRadiomics features
dc.subjectMedical image processing
dc.subjectMachine learning
dc.titleThe potential of radiomics features in the detection of hepatocellular carcinoma (HCC) in 2D liver MRI images by using machine learning methods
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu