Machine Learning Integration of MRI Intratumoral and Peritumoral Radiomics Features for Predicting PNSTs Postoperative Complications
Academic Radiology · 2026-05 · 33 (5) · 2211-2221
DOI: 10.1016/j.acra.2026.01.052 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academic Radiology, 2026-05. Verfasst von Jia Hao Liu und 5 weiteren Autor:innen. Beteiligte Einrichtung: Wound Repair Department, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China (J.H.L, F.Y.T., J.F.W., Y.H.C., R.W.H., J.W.); H…. Hauptthemen laut MEDLINE: Radiomics, Magnetic Resonance Imaging, Machine Learning, Postoperative Complications. Die Arbeit nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
RATIONALE AND OBJECTIVES: Postoperative complications (15-76%) substantially affect patients with peripheral nerve sheath tumors (PNSTs), yet objective preoperative risk tools are lacking. Magnetic resonance imaging (MRI) radiomics has focused mainly on intratumoral regions, while the predictive value of the peritumoral microenvironment remains unclear. MATERIALS AND METHODS: In this retrospective single-center study (Dec 2015-Jan 2024), 280 pathologically confirmed PNST patients with preoperative MRI were randomly split 8:2 into training (n=224) and test (n=56) cohorts. Intratumoral (Intra) and Peritumoral (Per; 2 mm expansion) regions were manually segmented and 1197 radiomic features extracted. Four models were constructed: Intra-model, Per-model, a fused Intra-model+Per-model region model (Imagefusion) and a concatenated Intra-model+Per-model feature model (intraPeri2mm). After reliability filtering, t-tests and least absolute shrinkage and selection operator selection, models were trained with machine-learning classifiers. Clinical predictors were assessed, and model performance evaluated using area under the receiver operating characteristic curve (AUC) and decision-curve analysis (DCA). RESULTS: Diabetes was the only independent clinical predictor, and the clinical model achieved a test AUC of 0.599. In the test cohort, both fusion models outperformed single-region models (intraPeri2mm AUC 0.899; Imagefusion AUC 0.895), with consistently greater net benefit on DCA. Incorporating diabetes yielded a small, nonsignificant gain for Imagefusion (Combined_1 AUC 0.917) and no further improvement for intraPeri2mm (Combined_2 AUC 0.889). All radiomics models significantly exceeded the clinical-only model (all p
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Wound Repair Department, The First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China (J.H.L, F.Y.T., J.F.W., Y.H.C., R.W.H., J.W.); Hand Surgery & Peripheral Nerve Surgery, The First Affiliated Hospital…
Zitieren
APA
Liu, J. H., Tang, F. Y., Wang, J. F., Cai, Y. H., Han, R. W., & Wang, J. (2026). Machine Learning Integration of MRI Intratumoral and Peritumoral Radiomics Features for Predicting PNSTs Postoperative Complications. Academic Radiology, 33(5), 2211-2221. https://doi.org/10.1016/j.acra.2026.01.052
Vancouver
Liu JH, Tang FY, Wang JF, Cai YH, Han RW, Wang J. Machine Learning Integration of MRI Intratumoral and Peritumoral Radiomics Features for Predicting PNSTs Postoperative Complications. Academic Radiology. 2026;33(5):2211-2221. doi:10.1016/j.acra.2026.01.052
BibTeX
@article{Liu2026,
author = {Liu, Jia Hao and Tang, Fang Ying and Wang, Ji Feng and Cai, Yun Hao and Han, Ren Wei and Wang, Jian},
title = {Machine Learning Integration of MRI Intratumoral and Peritumoral Radiomics Features for Predicting PNSTs Postoperative Complications},
journal = {Academic Radiology},
year = {2026},
volume = {33},
number = {5},
pages = {2211--2221},
doi = {10.1016/j.acra.2026.01.052}
}
EUVIMED