Share：
Share this content in WeChat
X
[Chinese] [PDF] 35 1
• Clinical Article •
Preoperative prediction of extra-pelvic peritoneal metastasis in epithelial ovarian cancer using habitat radiomics from T2-weighted MRI
WEI Mingxiang  WANG Xinyi  CHU Xiao  BAI Genji  CHEN Shuangqing 

Cite this article as: WEI M X, WANG X Y, CHU X, et al. Preoperative prediction of extra-pelvic peritoneal metastasis in epithelial ovarian cancer using habitat radiomics from T2-weighted MRI[J]. Chin J Magn Reson Imaging, 2026, 17(10): 117-124. DOI:10.12015/issn.1674-8034.2026.10.014.


[Abstract] Objective To investigate the value of a T2-weighted MRI-based habitat radiomics approach for accurate preoperative prediction of extra-pelvic peritoneal metastasis in patients with epithelial ovarian cancer (EOC).Materials and Methods Clinical and MRI data from 301 patients with pathologically confirmed EOC were retrospectively enrolled from two centers. The two centers served as a training cohort (n = 206) and an external test cohort (n = 95), respectively. Based on habitat analysis, tumor subregions were identified on preoperative axial fat-suppressed T2-weighted images, and radiomics features were extracted from each habitat subregion. A stable feature selection strategy based on 100 repeated random samplings was used, and the top 10 highly stable features were retained to construct the habitat radiomics model. The modeling performance of the logistic regression, support vector machine, random forest, and K-nearest neighbor algorithms was compared to determine the optimal model. SHapley Additive exPlanations (SHAP) was used to evaluate the contribution of each radiomics feature. Multivariable logistic regression was used to identify independent clinical predictors, which were combined with the habitat radiomics score to construct an ensemble model. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis. Differences in AUCs among models were compared using the DeLong test.Results All tumor regions were divided into three habitat subregions. The support vector machine model was identified as the optimal habitat radiomics model, with a mean AUC of 0.832 ± 0.086 in 10-fold cross-validation. After feature selection and SHAP analysis, H1_wavelet-HHL_glcm_MCC showed the highest reproducibility and made the greatest contribution to the prediction of the habitat radiomics model. Log-transformed serum carbohydrate antigen 125 (CA125) level was an independent clinical predictor of extra-pelvic peritoneal metastasis (P < 0.001). The ensemble model combining the habitat radiomics score and serum CA125 level achieved the best performance, with an AUC of 0.858 (95% CI: 0.765 to 0.935) in the external test cohort, which was significantly higher than that of the clinical model (P = 0.045).Conclusions The preoperative MRI-based habitat radiomics can be used to assess the risk of extra-pelvic peritoneal metastasis in EOC, and its combination with serum CA125 further improves predictive performance, potentially providing useful information for preoperative risk stratification, staging assessment, and treatment decision-making.
[Keywords] ovarian cancer；peritoneal metastasis；habitat imaging；radiomics；magnetic resonance imaging；T2-weighted imaging

WEI Mingxiang1   WANG Xinyi1   CHU Xiao1   BAI Genji2   CHEN Shuangqing1*  

1 Department of Radiology, The Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou Municipal Hospital, Gusu School, Nanjing Medical University, Suzhou 215000, China

2 Department of Radiology, The Affiliated Huaian No. 1 People's Hospital of Nanjing Medical University, Huaian 223300, China

Corresponding author: CHEN S Q, E-mail: sznaonao@163.com

Conflicts of interest   None.

Received  2026-06-11
Accepted  2026-09-19
DOI: 10.12015/issn.1674-8034.2026.10.014
Cite this article as: WEI M X, WANG X Y, CHU X, et al. Preoperative prediction of extra-pelvic peritoneal metastasis in epithelial ovarian cancer using habitat radiomics from T2-weighted MRI[J]. Chin J Magn Reson Imaging, 2026, 17(10): 117-124. DOI:10.12015/issn.1674-8034.2026.10.014.

[1]
SUNG H, FERLAY J, SIEGEL R L, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249. DOI: 10.3322/caac.21660.
[2]
SIEGEL R L, KRATZER T B, WAGLE N S, et al. Cancer statistics, 2026[J/OL]. CA Cancer J Clin, 2026, 76: e70043 [2026-06-10]. https://acsjournals.onlinelibrary.wiley.com/doi/10.3322/caac.70043. DOI: 10.3322/caac.70043.
[3]
SCOTT C L, BANERJEE S, JOLY F, et al. Ovarian cancer[J/OL]. Nat Rev Dis Primers, 2026, 12(1): 10 [2026-06-10]. https://www.nature.com/articles/s41572-026-00686-x. DOI: 10.1038/s41572-026-00686-x.
[4]
MEI S S, CHEN X, WANG K, et al. Tumor microenvironment in ovarian cancer peritoneal metastasis[J/OL]. Cancer Cell Int, 2023, 23: 11 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/36698173/. DOI: 10.1186/s12935-023-02854-5.
[5]
RENZ M, FRIEDLANDER M, BEREK J S. Cancer of the ovary, fallopian tube, and peritoneum: 2025 update[J]. Int J Gynaecol Obstet, 2025, 171(Suppl 1): 6-35. DOI: 10.1002/ijgo.70282.
[6]
GONZÁLEZ-MARTÍN A, HARTER P, LEARY A, et al. Newly diagnosed and relapsed epithelial ovarian cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up[J]. Ann Oncol, 2023, 34(10): 833-848. DOI: 10.1016/j.annonc.2023.07.011.
[7]
DU BOIS A, BAERT T, VERGOTE I. Role of neoadjuvant chemotherapy in advanced epithelial ovarian cancer[J]. J Clin Oncol, 2019, 37(27): 2398-2405. DOI: 10.1200/jco.19.00022.
[8]
LHEUREUX S, GOURLEY C, VERGOTE I, et al. Epithelial ovarian cancer[J]. Lancet, 2019, 393(10177): 1240-1253. DOI: 10.1016/S0140-6736(18)32552-2.
[9]
RIZZO S, DEL GRANDE M, MANGANARO L, et al. Imaging before cytoreductive surgery in advanced ovarian cancer patients[J]. Int J Gynecol Cancer, 2020, 30(1): 133-138. DOI: 10.1136/ijgc-2019-000819.
[10]
GARCIA PRADO J, GONZÁLEZ HERNANDO C, VARILLAS DELGADO D, et al. Diffusion-weighted magnetic resonance imaging in peritoneal carcinomatosis from suspected ovarian cancer: Diagnostic performance in correlation with surgical findings[J/OL]. Eur J Radiol, 2019, 121: 108696 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/31683251/. DOI: 10.1016/j.ejrad.2019.108696.
[11]
POWER J W, DEMPSEY P J, YATES A, et al. Peritoneal malignancy: anatomy, pathophysiology and an update on modern day imaging[J/OL]. Br J Radiol, 2022, 95(1132): 20210217 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/34826229/. DOI: 10.1259/bjr.20210217.
[12]
LAMBIN P, LEIJENAAR R T H, DEIST T M, et al. Radiomics: the bridge between medical imaging and personalized medicine[J]. Nat Rev Clin Oncol, 2017, 14(12): 749-762. DOI: 10.1038/nrclinonc.2017.141.
[13]
SONG X L, REN J L, YAO T Y, et al. Radiomics based on multisequence magnetic resonance imaging for the preoperative prediction of peritoneal metastasis in ovarian cancer[J]. Eur Radiol, 2021, 31(11): 8438-8446. DOI: 10.1007/s00330-021-08004-7.
[14]
YU X Y, REN J L, JIA Y S, et al. Multiparameter MRI radiomics model predicts preoperative peritoneal carcinomatosis in ovarian cancer[J/OL]. Front Oncol, 2021, 11: 765652 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/34790579/. DOI: 10.3389/fonc.2021.765652.
[15]
YANG Z T, WU H, GAO H Y, et al. Progress in the application of habitat imaging in multi-system tumors[J]. Chin J Magn Reson Imaging, 2025, 16(3): 222-227. DOI: 10.12015/issn.1674-8034.2025.03.038.
[16]
WU J J, XIA Y W, WANG X C, et al. Radiomics++: review of habitat imaging analysis for decoding tumor heterogeneity[J]. Annu Rev Biomed Eng, 2026, 28(1): 219-248. DOI: 10.1146/annurev-bioeng-031825-040442.
[17]
BI Q, MIAO K, XU N, et al. Habitat radiomics based on MRI for predicting platinum resistance in patients with high-grade serous ovarian carcinoma: a multicenter study[J]. Acad Radiol, 2024, 31(6): 2367-2380. DOI: 10.1016/j.acra.2023.11.038.
[18]
LIN Z J, CHEN X F, YANG J, et al. Habitat radiomics based on diffusion-weighted MRI for predicting PARP inhibitor resistance in advanced high-grade serous ovarian cancer[J]. Acad Radiol, 2026, 33(8): 3402-3413. DOI: 10.1016/j.acra.2026.04.011.
[19]
ZWANENBURG A, VALLIÈRES M, ABDALAH M A, et al. The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping[J]. Radiology, 2020, 295(2): 328-338. DOI: 10.1148/radiol.2020191145.
[20]
HUANG Y Y, PENG W R, YANG H, et al. Radiomics and radiogenomics in ovarian cancer: a review with a focus on ultrasound applications[J/OL]. Cancer Imaging, 2025, 25(1): 130 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/41220029/. DOI: 10.1186/s40644-025-00950-5.
[21]
LIN L L, WU H W, WANG C, et al. Spatial habitats radiomics from multiparametric magnetic resonance imaging predict platinum resistance in epithelial ovarian cancer[J/OL]. J Ovarian Res, 2026, 19(1): 201 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/42015267/. DOI: 10.1186/s13048-026-02097-8.
[22]
O'SHEA R, HORST C, MANICKAVASAGAR T, et al. Independent testing of published CT models for PD-L1 status in non-small cell lung cancer[J/OL]. Radiology, 2026, 318(3): e241962 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/41914868/. DOI: 10.1148/radiol.241962.
[23]
WANG X Y, QUAN T, CHU X, et al. Deep learning radiomics nomogram based on MRI for differentiating between borderline ovarian tumors and stage I ovarian cancer: a multicenter study[J]. Acad Radiol, 2025, 32(6): 3485-3497. DOI: 10.1016/j.acra.2024.12.067.
[24]
KOCAK B, BAESSLER B, BAKAS S, et al. CheckList for EvaluAtion of Radiomics research (CLEAR): a step-by-step reporting guideline for authors and reviewers endorsed by ESR and EuSoMII[J/OL]. Insights Imaging, 2023, 14(1): 75 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/37142815/. DOI: 10.1186/s13244-023-01415-8.
[25]
TEJANI A S, KLONTZAS M E, GATTI A A, et al. Checklist for artificial intelligence in medical imaging (CLAIM): 2024 update[J/OL]. Radiol Artif Intell, 2024, 6(4): e240300 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/38809149/. DOI: 10.1148/ryai.240300.
[26]
HU Y, LIU Y, ZHENG Y N, et al. Application of different radiomics models based on MRI conventional T2WI in preoperative tri-classification of ovarian epithelial tumors[J]. Chin J Magn Reson Imaging, 2021, 12(12): 34-38, 54. DOI: 10.12015/issn.1674-8034.2021.12.007.
[27]
WEI M X, FENG G N, WANG X Y, et al. Deep learning radiomics nomogram based on magnetic resonance imaging for differentiating type I/II epithelial ovarian cancer[J]. Acad Radiol, 2024, 31(6): 2391-2401. DOI: 10.1016/j.acra.2023.08.002.
[28]
YUAN L, ZHANG J L, MA L N, et al. Non-invasive quantitative visualization of multi-parametric MRI habitat imaging for predicting prostate cancer risk degree[J]. Chin J Radiol, 2025, 59(4): 393-400. DOI: 10.3760/cma.j.cn112149-20240624-00348.
[29]
CHEN W H, YU Y X, GU W H, et al. The value of Gd-EOB-DTPA-enhanced MRI habitat radiomic features in predicting CK19 expression and prognosis of hepatocellular carcinoma[J]. Chin J Radiol, 2025, 59(11): 1275-1285. DOI: 10.3760/cma.j.cn112149-20250525-00281.
[30]
ZHANG M H, CHENG S S, JIN Y, et al. Roles of CA125 in diagnosis, prediction, and oncogenesis of ovarian cancer[J/OL]. Biochim Biophys Acta Rev Cancer, 2021, 1875(2): 188503 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/33421585/. DOI: 10.1016/j.bbcan.2021.188503.
[31]
ZHAO J J, QIN R H, CHEN H, et al. A nomogram based on glycomic biomarkers in serum and clinicopathological characteristics for evaluating the risk of peritoneal metastasis in gastric cancer[J/OL]. Clin Proteomics, 2020, 17: 34 [2026-06-10]. https://pubmed.ncbi.nlm.nih.gov/32968368/. DOI: 10.1186/s12014-020-09297-4.
[32]
XUE B H, JIANG J, CHEN L, et al. Development and validation of a radiomics model based on 18F-FDG PET of primary gastric cancer for predicting peritoneal metastasis[J/OL]. Front Oncol, 2021, 11: 740111 [2026-06-10]. https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.740111. DOI: 10.3389/fonc.2021.740111.

PREV The value of IVIM-DWI combined with clinical parameters in predicting disease-free survival of endometrial cancer
NEXT The value of radiomics nomogram based on multiparametric MRI in differentiating benign and malignant ovarian O-RADS 4 masses
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn