Differentiation of cervical cancer subtypes using machine learning models on MRI images
Polish Journal of Radiology · 22.04.2026 · 91 (1) · 195-204
DOI: 10.5114/pjr/216134 ↗Kurz gefasst
Fachartikel in der Zeitschrift Polish Journal of Radiology, online erschienen am 22.04.2026. Verfasst von Asiye Sozeri und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Karabük Training and Research Hospital, Karabuk, Turkey. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Abstract (Originalsprache)
Purpose
To develop and compare various radiomics-based machine learning (ML) models for distinguishing cervical squamous cell carcinoma (SCC) from non-SCC histopathological subtypes, using multiparametric magnetic resonance imaging (MRI) and clinical data. Material and methods This retrospective study included 88 women (mean age, 51.1 ±13.0 years; range, 25-83 years) with histopathologically confirmed cervical cancer (47 SCC, 41 non-SCC). For each patient, axial and sagittal T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) were available, along with clinical metadata. Radiomic features (shape, first-order, and texture features including gray-level co-occurrence matrix, gray-level run length matrix, and gray-level size zone matrix) were extracted from each sequence, yielding 945 features per patient. A feature set was created by combining features from all three sequences. The least absolute shrinkage and selection operator (LASSO) method was used for feature selection. Three supervised classifiers – random forest (RF), support vector machine (SVM), and logistic regression (LR) – were trained to classify SCC versus non-SCC.
Results
Among the single-sequence models, the sagittal T2WI sequence demonstrated the strongest performance, achieving receiver operating characteristic (ROC) area under the curve (AUC) values of 0.839 with SVM and 0.732 with LR. In contrast, the axial T2WI RF model showed poor discriminative ability (ROC-AUC: 0.464), indicating that this sequence alone provides limited value for subtype differentiation despite showing acceptable accuracy metrics. The combined multi-sequence model yielded the highest overall performance. Using LASSO-selected features, RF achieved the best ROC-AUC (0.958), followed closely by SVM (0.953) and LR (0.951), with all models attaining accuracies above 0.91.
Conclusions
Feature-level integration of axial T2WI, sagittal T2WI, and DWI substantially enhances ML performance in differentiating SCC from non-SCC cervical cancer subtypes. Compared with single-sequence models, the combined multi-sequence approach provides more robust and well-balanced classification, underscoring the complementary value of multiparametric MRI for histopathological subtype prediction.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, Karabük Training and Research Hospital, Karabuk, Turkey.
- Computer Engineering Department, College of Engineering, Koc University, Istanbul, Turkey.
- Department of Gynecologic Oncology, Ankara Bilkent City Hospital, Ankara, Turkey.
- Department of Radiology, Ankara Bilkent City Hospital, Ankara, Turkey.
Zitieren
APA
Sozeri, A., Gokmen, N., Kayali, E., Cuylan, Z. F., Eroglu, F., Parlak, I. S., & Comert, G. K. (2026). Differentiation of cervical cancer subtypes using machine learning models on MRI images. Polish Journal of Radiology, 91(1), 195-204. https://doi.org/10.5114/pjr/216134
Vancouver
Sozeri A, Gokmen N, Kayali E, Cuylan ZF, Eroglu F, Parlak IS, et al. Differentiation of cervical cancer subtypes using machine learning models on MRI images. Polish Journal of Radiology. 2026;91(1):195-204. doi:10.5114/pjr/216134
BibTeX
@article{Sozeri2026,
author = {Sozeri, Asiye and Gokmen, Neslihan and Kayali, Eda and Cuylan, Zeliha Firat and Eroglu, Fatih and Parlak, Izzet Selcuk and Comert, Gunsu Kimyon},
title = {Differentiation of cervical cancer subtypes using machine learning models on MRI images},
journal = {Polish Journal of Radiology},
year = {2026},
volume = {91},
number = {1},
pages = {195--204},
doi = {10.5114/pjr/216134}
}
EUVIMED