Share：
Share this content in WeChat
X
[Chinese] [PDF] 8 1
• Clinical Article •
Interpretable machine learning model based on DCE-MRI habitat imaging radiomics for predicting lymph node metastasis in rectal cancer
SUN Yun  LI Feixiang  CHEN Xuemin  ZHANG Yingying  HUANG Gang 

DOI:10.12015/issn.1674-8034.2026.01.010.


[Abstract] Objective To construct an interpretable integrated model based on dynamic contrast-enhanced MRI (DCE-MRI) habitat imaging radiomics and clinical features, and to assess its utility in predicting lymph node metastasis (LNM) status in rectal cancer.Materials and Methods A retrospective analysis was conducted on the clinicopathological and imaging data of 148 patients with rectal cancer admitted to Gansu Provincial People's Hospital between January 2016 and July 2024. Patients were stratified into LNM-positive and LNM-negative groups based on postoperative pathological confirmation. They were then randomly divided into a training cohort (n = 103) and a test cohort (n = 45) in a 7∶3 ratio. The region of interest (ROI) was manually delineated on the DCE-MRI parametric map using ITK-SNAP software. Subsequently, 19 standardized radiomics features were extracted from the Ktrans maps. K-means clustering (K = 4) was applied to partition the tumor into distinct habitat subregions. Radiomics features were extracted separately from each tumor subregion (habitat-specific features) and from the whole tumor volume (whole-tumor features). The intra-class correlation coefficient (ICC) was calculated to assess the reproducibility of the whole-tumor radiomics feature extraction. Feature selection involved Z-score normalization, correlation analysis, and the least absolute shrinkage and selection operator (LASSO) algorithm. Predictive models for LNM status were developed using four machine learning classifiers: extremely randomized trees, logistic regression, random forest, and support vector machine. These models were built based on habitat-specific radiomics features and whole-tumor radiomics features separately. Logistic regression was also used to identify independent clinical predictors and construct a clinical model. Finally, an integrated model was built by combining significant clinical predictors with the radiomics signature derived from the habitat analysis. Model performance was evaluated using the receiver operating characteristic (ROC) curve and quantified by the area under the curve (AUC). Decision curve analysis (DCA) was performed to assess the clinical utility of the models. The importance of features in the final integrated model was determined, and the model's predictions were explained visually using Shapley additive explanations (SHAP) analysis.Result Univariate analysis identified carcinoembryonic antigen (CEA) level and MRI-reported N-stage as significant predictors of lymph node status in the training cohort [odds ratios (OR) = 2.346 and 7.727, respectively; 95% confidence intervals (CI): 1.052 to 5.233 and 2.273 to 26.268, respectively; P < 0.05]. The predictive model based on habitat radiomics features demonstrated superior performance, with AUC values of 0.890 (training cohort) and 0.801 (test cohort), outperforming the whole-tumor radiomics model (AUC: 0.774 training, 0.684 test). The integrated model, combining clinical features with the habitat radiomics signature, achieved the highest AUC values: 0.896 in the training cohort and 0.866 in the test cohort. DCA indicated that the integrated model provided a higher net clinical benefit across a range of threshold probabilities. SHAP analysis provided quantitative interpretability for the integrated model's predictions, revealing the habitat radiomics score as the most significant predictor.Conclusions The interpretable integrated model, constructed using preoperative DCE-MRI habitat imaging radiomics features and clinical factors, accurately predicts lymph node status in rectal cancer patients. By providing visual interpretation of individual predictions through SHAP, this model offers a valuable tool to support personalized treatment decision-making.
[Keywords] rectal cancer；lymph node metastasis；magnetic resonance imaging；radiomics；habitat analysis

SUN Yun1   LI Feixiang1   CHEN Xuemin1   ZHANG Yingying1   HUANG Gang2*  

1 The First Clinical Medical College of Gansu University of Chinese Medicine, Lanzhou, 730000, China

2 Department of Radiology, Gansu Provincial Hospital, Lanzhou, 730000, China

Corresponding author: HUANG G, E-mail: huang_g2024@163.com

Conflicts of interest   None.

Received  2025-07-28
Accepted  2025-12-19
DOI: 10.12015/issn.1674-8034.2026.01.010
DOI:10.12015/issn.1674-8034.2026.01.010.

[1]
BRAY F, LAVERSANNE M, SUNG H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2024, 74(3): 229-263. DOI: 10.3322/caac.21834.
[2]
General Office of National Health Commission of the People's Republic of China, Oncology Branch of the Chinese Medical Association. The Standard for Diagnosis and Treatment of Chinese Colorectal Cancer (2023 Version)[J]. Med J Peking Union Med Coll Hosp, 2023, 14(4): 706-733. DOI: 10.12290/xhyxzz.2023-0315.
[3]
JOUNG J G, OH B Y, HONG H K, et al. Tumor heterogeneity predicts metastatic potential in colorectal cancer[J]. Clin Cancer Res, 2017, 23(23): 7209-7216. DOI: 10.1158/1078-0432.CCR-17-0306.
[4]
BENSON A B, VENOOK A P, AL-HAWARY M M, et al. Rectal cancer, version 2.2022, NCCN clinical practice guidelines in oncology[J]. J Natl Compr Canc Netw, 2022, 20(10): 1139-1167. DOI: 10.6004/jnccn.2022.0051.
[5]
BENSON A B, VENOOK A P, ADAM M, et al. NCCN guidelines® insights: rectal cancer, version 3.2024[J]. J Natl Compr Canc Netw, 2024, 22(6): 366-375. DOI: 10.6004/jnccn.2024.0041.
[6]
ABBASPOUR E, KARIMZADHAGH S, MONSEF A, et al. Application of radiomics for preoperative prediction of lymph node metastasis in colorectal cancer: a systematic review and meta-analysis[J]. Int J Surg, 2024, 110(6): 3795-3813. DOI: 10.1097/JS9.0000000000001239.
[7]
GRÖNE J, LOCH F N, TAUPITZ M, et al. Accuracy of various lymph node staging criteria in rectal cancer with magnetic resonance imaging[J]. J Gastrointest Surg, 2018, 22(1): 146-153. DOI: 10.1007/s11605-017-3568-x.
[8]
HAO Y F, ZHENG J Y, LI W Q, et al. Ultra-high b-value DWI in rectal cancer: image quality assessment and regional lymph node prediction based on radiomics[J]. Eur Radiol, 2025, 35(1): 49-60. DOI: 10.1007/s00330-024-10958-3.
[9]
QIN S Y, LIU K, CHEN Y Y, et al. Prediction of pathological response and lymph node metastasis after neoadjuvant therapy in rectal cancer through tumor and mesorectal MRI radiomic features[J/OL]. Sci Rep, 2024, 14(1): 21927 [2025-10-30]. https://pubmed.ncbi.nlm.nih.gov/39304726/. DOI: 10.1038/s41598-024-72916-9.
[10]
ZHENG Y F, CHEN X, ZHANG H, et al. Multiparametric MRI-based radiomics nomogram for the preoperative prediction of lymph node metastasis in rectal cancer: a two-center study[J/OL]. Eur J Radiol, 2024, 178: 111591 [2025-03-05]. https://pubmed.ncbi.nlm.nih.gov/39013271/. DOI: 10.1016/j.ejrad.2024.111591.
[11]
WEI Q R, CHEN L, HOU X Y, et al. Multiparametric MRI-based radiomic model for predicting lymph node metastasis after neoadjuvant chemoradiotherapy in locally advanced rectal cancer[J/OL]. Insights Imaging, 2024, 15(1): 163 [2025-09-07]. https://pubmed.ncbi.nlm.nih.gov/38922456/. DOI: 10.1186/s13244-024-01726-4.
[12]
O'CONNOR J P B, ROSE C J, WATERTON J C, et al. Imaging intratumor heterogeneity: role in therapy response, resistance, and clinical outcome[J]. Clin Cancer Res, 2015, 21(2): 249-257. DOI: 10.1158/1078-0432.CCR-14-0990.
[13]
YANG Y J, XU Z Y, CAI Z P, et al. Novel deep learning radiomics nomogram-based multiparametric MRI for predicting the lymph node metastasis in rectal cancer: A dual-center study[J/OL]. J Cancer Res Clin Oncol, 2024, 150(10): 450 [2025-02-13]. https://pubmed.ncbi.nlm.nih.gov/39379733/. DOI: 10.1007/s00432-024-05986-x.
[14]
SUN Y, LU Z X, YANG H J, et al. Prediction of lateral lymph node metastasis in rectal cancer patients based on MRI using clinical, deep transfer learning, radiomic, and fusion models[J/OL]. Front Oncol, 2024, 14: 1433190 [2025-09-02]. https://pubmed.ncbi.nlm.nih.gov/39099685/. DOI: 10.3389/fonc.2024.1433190.
[15]
AO W Q, WU S K, WANG N, et al. Novel deep learning algorithm based MRI radiomics for predicting lymph node metastases in rectal cancer[J/OL]. Sci Rep, 2025, 15(1): 12089 [2025-10-03]. https://pubmed.ncbi.nlm.nih.gov/40204902/. DOI: 10.1038/s41598-025-96618-y.
[16]
ZHAO J Y, ZHENG P, XU T, et al. Preoperative assessment of lymph node metastasis in rectal cancer using deep learning: investigating the utility of various MRI sequences[J]. Ann Surg Oncol, 2025, 32(10): 7304-7314. DOI: 10.1245/s10434-025-17717-8.
[17]
OUCHI A, IWAHORI Y, SUZUKI K, et al. Artificial intelligence imaging diagnosis using super-resolution and three-dimensional shape for lymph node metastasis of low rectal cancer: a pilot study from a single center[J]. Dis Colon Rectum, 2024, 67(9): 1131-1138. DOI: 10.1097/DCR.0000000000003381.
[18]
KIM M, PARK J E, KIM H S, et al. Spatiotemporal habitats from multiparametric physiologic MRI distinguish tumor progression from treatment-related change in post-treatment glioblastoma[J]. Eur Radiol, 2021, 31(8): 6374-6383. DOI: 10.1007/s00330-021-07718-y.
[19]
DEXTRAZE K, SAHA A, KIM D, et al. Spatial habitats from multiparametric MR imaging are associated with signaling pathway activities and survival in glioblastoma[J]. Oncotarget, 2017, 8(68): 112992-113001. DOI: 10.18632/oncotarget.22947.
[20]
MOON H H, PARK J E, KIM N, et al. Prospective longitudinal analysis of imaging-based spatiotemporal tumor habitats in glioblastoma, IDH-wild type: implication in patient outcome using multiparametric physiologic MRI[J/OL]. BMC Cancer, 2024, 24(1): 1197 [2025-07-03]. https://pubmed.ncbi.nlm.nih.gov/39334005/. DOI: 10.1186/s12885-024-12939-7.
[21]
CHEN X W, JIANG H, PAN M, et al. Habitat radiomics predicts occult lymph node metastasis and uncovers immune microenvironment of head and neck cancer[J/OL]. J Transl Med, 2025, 23(1): 498 [2025-09-15]. https://pubmed.ncbi.nlm.nih.gov/40312725/. DOI: 10.1186/s12967-025-06474-7.
[22]
JIA P F, LI Y Y, LI H N, et al. Habitat radiomics assists radiologists in accurately diagnosing lymph node metastasis of adenocarcinoma of the esophagogastric junction[J/OL]. Insights Imaging, 2025, 16(1): 90 [2025-06-12]. https://pubmed.ncbi.nlm.nih.gov/40272618/. DOI: 10.1186/s13244-025-01969-9.
[23]
LI F X, SUN Y, GONG J T, et al. Study on predicting LVSI status in preoperative cervical cancer patients without lymph node metastasis using habitat radiomics based on DCE-MRI quantitative parametric maps[J]. Chin J Magn Reson Imag, 2025, 16(10): 89-97. DOI: 10.12015/issn.1674-8034.2025.10.014.
[24]
AMIN M B, GREENE F L, EDGE S B, et al. The eighth edition AJCC cancer staging manual: continuing to build a bridge from a population-based to a more "personalized" approach to cancer staging[J]. CA Cancer J Clin, 2017, 67(2): 93-99. DOI: 10.3322/caac.21388.
[25]
HUANG Y Q, LIANG C H, HE L, et al. Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer[J]. J Clin Oncol, 2016, 34(18): 2157-2164. DOI: 10.1200/JCO.2015.65.9128.
[26]
LI M L, ZHANG J, DAN Y B, et al. A clinical-radiomics nomogram for the preoperative prediction of lymph node metastasis in colorectal cancer[J/OL]. J Transl Med, 2020, 18(1): 46 [2024-12-16]. https://pubmed.ncbi.nlm.nih.gov/32000813/. DOI: 10.1186/s12967-020-02215-0.
[27]
ZENG D X, YANG Z, TAN L, et al. Risk factors for lateral pelvic lymph node metastasis in patients with lower rectal cancer: a systematic review and meta-analysis[J/OL]. Front Oncol, 2023, 13: 1219608 [2024-09-21]. https://pubmed.ncbi.nlm.nih.gov/37746256/. DOI: 10.3389/fonc.2023.1219608.
[28]
WU H M, WANG Y Z, DENG M, et al. Preoperative inflammatory markers and tumor markers in predicting lymphatic metastasis and postoperative complications in colorectal cancer: a retrospective study[J/OL]. BMC Surg, 2025, 25(1): 71 [2025-09-27]. https://pubmed.ncbi.nlm.nih.gov/39966788/. DOI: 10.1186/s12893-025-02795-y.
[29]
LUO H, GOU Y Q, WANG Y S, et al. A novel pN stage prediction model for resectable rectal adenocarcinoma based on preoperative MRI features and multiregional apparent diffusion coefficients[J]. Eur Radiol, 2025, 35(9): 5760-5770. DOI: 10.1007/s00330-025-11528-x.
[30]
BROUWER N P M, STIJNS R C H, LEMMENS V E P P, et al. Clinical lymph node staging in colorectal cancer; a flip of the coin [J]. Eur J Surg Oncol, 2018, 44(8): 1241-1246. DOI: 10.1016/j.ejso.2018.04.008.
[31]
KELLER D S, BERHO M, PEREZ R O, et al. The multidisciplinary management of rectal cancer[J]. Nat Rev Gastroenterol Hepatol, 2020, 17(7): 414-429. DOI: 10.1038/s41575-020-0275-y.
[32]
QU W N, WANG J, HU X M, et al. MRI radiomic study on prediction of nonenlarged lymph node metastasis of rectal cancer: reduced field-of-view versus conventional DWI[J/OL]. Eur Radiol Exp, 2025, 9(1): 34 [2025-04-23]. https://pubmed.ncbi.nlm.nih.gov/40120024/. DOI: 10.1186/s41747-025-00575-0.
[33]
WANG Y Y, XIE B, WANG K, et al. Multi-parametric MRI habitat radiomics based on interpretable machine learning for preoperative assessment of microsatellite instability in rectal cancer[J]. Acad Radiol, 2025, 32(7): 3975-3988. DOI: 10.1016/j.acra.2025.02.009.
[34]
SHI S M, JIANG T, LIU H, et al. Habitat radiomics based on MRI for predicting metachronous liver metastasis in locally advanced rectal cancer: a two-center study[J]. Acad Radiol, 2025, 32(6): 3370-3383. DOI: 10.1016/j.acra.2025.02.046.
[35]
WU J, CAO G H, SUN X L, et al. Intratumoral spatial heterogeneity at perfusion MR imaging predicts recurrence-free survival in locally advanced breast cancer treated with neoadjuvant chemotherapy[J]. Radiology, 2018, 288(1): 26-35. DOI: 10.1148/radiol.2018172462.
[36]
GILLIES R J, BALAGURUNATHAN Y. Perfusion MR imaging of breast cancer: insights using "habitat imaging"[J]. Radiology, 2018, 288(1): 36-37. DOI: 10.1148/radiol.2018180271.
[37]
LI X W, HU Y W, XIE Y X, et al. Whole-tumor histogram analysis of diffusion-weighted imaging and dynamic contrast-enhanced MRI for soft tissue sarcoma: correlation with HIF-1alpha expression[J]. Eur Radiol, 2023, 33(6): 3961-3973. DOI: 10.1007/s00330-022-09296-z.
[38]
SACHANI P, DHANDE R, PARIHAR P, et al. Enhancing the understanding of breast vascularity through insights from dynamic contrast-enhanced magnetic resonance imaging: A comprehensive review[J/OL]. Cureus, 2024, 16(9): e70226 [2025-06-07]. https://pubmed.ncbi.nlm.nih.gov/39463566/. DOI: 10.7759/cureus.70226.
[39]
CHEN Y D, JIANG Z J, GUAN X C, et al. The value of multi-parameter diffusion and perfusion magnetic resonance imaging for evaluating epithelial-mesenchymal transition in rectal cancer[J/OL]. Eur J Radiol, 2022, 150: 110245 [2024-08-13]. https://pubmed.ncbi.nlm.nih.gov/35299114/. DOI: 10.1016/j.ejrad.2022.110245.
[40]
LI C L, YIN J D. Radiomics based on T2-weighted imaging and apparent diffusion coefficient images for preoperative evaluation of lymph node metastasis in rectal cancer patients[J/OL]. Front Oncol, 2021, 11: 671354 [2024-12-23]. https://pubmed.ncbi.nlm.nih.gov/34041033/. DOI: 10.3389/fonc.2021.671354.
[41]
DONG X, REN G, CHEN Y H, et al. Effects of MRI radiomics combined with clinical data in evaluating lymph node metastasis in mrT1-3a staging rectal cancer[J/OL]. Front Oncol, 2023, 13: 1194120 [2025-07-26]. https://pubmed.ncbi.nlm.nih.gov/37909021/. DOI: 10.3389/fonc.2023.1194120.
[42]
LI R, SHINDE A, LIU A, et al. Machine learning-based interpretation and visualization of nonlinear interactions in prostate cancer survival[J]. JCO Clin Cancer Inform, 2020, 4: 637-646. DOI: 10.1200/CCI.20.00002.

PREV Prediction of treatment response to neoadjuvant chemoradiotherapy in patients with locally advanced rectal cancer by interpretable model based on multiparametric MRI
NEXT Prediction of lymph node metastasis in cervical cancer using virtual magnetic resonance elastography
  


LINK


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