A Predictive MRI Radiomics Model for Histologic Differentiation in Soft Tissue Sarcomas
Cancers · 21.05.2026 · 18 (10) · 1667
DOI: 10.3390/cancers18101667 ↗Kurz gefasst
Fachartikel in der Zeitschrift Cancers, online erschienen am 21.05.2026. Verfasst von Laetitia Perronne und 16 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USA. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Department of Radiology, Northwestern University Feinberg School of Medicine.
Abstract (Originalsprache)
Background/Objectives: The aim of this study was to develop and validate a robust, radiomics-based classification model that uses pre-treatment MRI to non-invasively differentiate among major soft tissue sarcoma (STS) subtypes and a benign mimic.
Methods
In this retrospective study, a cohort of 332 patients with biopsy-proven leiomyosarcoma, myxofibrosarcoma, myxoid liposarcoma, dedifferentiated liposarcoma, and undifferentiated pleomorphic sarcoma, along with the benign mimic intramuscular myxoma, was analyzed. Pre-treatment T1-weighted fat-saturated contrast-enhanced and T2-weighted fat-saturated MRI sequences were used for analysis. Following manual tumor segmentation, 1240 three-dimensional radiomic features were extracted. An XGBoost classifier was trained and validated using a robust 250-iteration bootstrap framework with nested cross-validation to ensure rigorous feature selection and unbiased performance evaluation. The model’s performance was assessed independently on T1-only, T2-only, and combined T1+T2 feature sets.
Results
The combined T1 and T2 model achieved superior performance with an accuracy of 0.68 ± 0.04 and an AUC of 0.92 ± 0.02. At the subtype level, balanced accuracy was highest for intramuscular myxoma (0.91 ± 0.05), dedifferentiated liposarcoma (0.84 ± 0.06), and leiomyosarcoma (0.83 ± 0.05). SHAP analysis identified key features driving predictions, such as low T2 GLSZM Zone Size Entropy for myxoma and high T2 GLSZM Gray-Level Variance for leiomyosarcoma, which aligns with known pathological characteristics. Misclassifications predominantly occurred between subtypes with overlapping radiomic profiles.
Conclusions
Radiomics applied to pre-treatment MRI enables robust, non-invasive classification of STS subtypes, demonstrating strong clinical potential for improving diagnostic confidence and informing triage strategies.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USA.
- Robert H. Lurie Comprehensive Cancer Center, Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USA.
- Department of Surgery, Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USA.
- Department of Pathology, Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USA.
- Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USA.
ORCID-Profile: Laetitia Perronne, Nicolò Gennaro, Zuzanna Kobus, Hatice Savas, Gorkem Durak, Seth Pollack, Yuri S Velichko
Förderung
- Department of Radiology, Northwestern University Feinberg School of Medicine
Zitieren
APA
Perronne, L., Gennaro, N., Kobus, Z., Seo, M., Borhani, A. A., Kelahan, L., Savas, H., Avery, R., Subedi, K., Krumpelman, C., Durak, G., Bagci, U., Chawla, A., Alexiev, B., Viveiros, P. H., Pollack, S., & Velichko, Y. S. (2026). A Predictive MRI Radiomics Model for Histologic Differentiation in Soft Tissue Sarcomas. Cancers, 18(10), 1667. https://doi.org/10.3390/cancers18101667
Vancouver
Perronne L, Gennaro N, Kobus Z, Seo M, Borhani AA, Kelahan L, et al. A Predictive MRI Radiomics Model for Histologic Differentiation in Soft Tissue Sarcomas. Cancers. 2026;18(10):1667. doi:10.3390/cancers18101667
BibTeX
@article{Perronne2026,
author = {Perronne, Laetitia and Gennaro, Nicolò and Kobus, Zuzanna and Seo, Mirinae and Borhani, Amir A and Kelahan, Linda and Savas, Hatice and Avery, Ryan and Subedi, Kamal and Krumpelman, Chase and Durak, Gorkem and Bagci, Ulas and Chawla, Akhil and Alexiev, Borislav and Viveiros, Pedro Hermida de and Pollack, Seth and Velichko, Yuri S},
title = {A Predictive MRI Radiomics Model for Histologic Differentiation in Soft Tissue Sarcomas},
journal = {Cancers},
year = {2026},
volume = {18},
number = {10},
pages = {1667},
doi = {10.3390/cancers18101667}
}
EUVIMED