Share：
Share this content in WeChat
X
[Chinese][PDF]711
• Clinical Article •
Predictive value of machine learning model based on ADC radiomics in evaluating the invasion depth of endometrial carcinoma
CUI Jing  GUO Ran  XIN Ruiqiang 

Cite this article as: CUI J, GUO R, XIN R Q. Predictive value of machine learning model based on ADC radiomics in evaluating the invasion depth of endometrial carcinoma[J]. Chin J Magn Reson Imaging, 2025, 16(3): 77-82. DOI:10.12015/issn.1674-8034.2025.03.012.


[Abstract] Objective To explore the predictive value of radiomics models based on apparent diffusion coefficient (ADC) in evaluating the myometrial invasion depth of endometrial carcinoma (EC), providing a reliable evidence for clinicians to formulate treatment plans.Materials and Methods Retrospective analysis of 155 patients with EC who underwent preoperative pelvic MR examination and were confirmed by pathology after operation from January 2016 to December 2023 in Beijing Luhe Hospital (superficial myometrial invasion = 114, deep invasion = 41), and randomly divided into training set (n = 124) and validation set (n = 31) in a 4∶1 ratio. The ITK-SNAP software was used to delineate the tumor regions layer by layer on the ADC maps, and the radiomics features were extracted, the extracted features were normalized. Pearson correlation coefficients (PCC) and least absolute shrinkage and selection operator (LASSO) were used to reduce features dimensionality, and the importance of the screened radiomics features was ranked according to the weight coefficient, the top 10 features were used to build radiomics models using three algorithms: logistic regression (LR), random forest (RF), and gradient boosting machine (GBM). The models were validated on the validation set. The performance of three radiomics models were evaluated by the receiver operating characteristic (ROC) curve, calibration curves, and decision curve analysis (DCA). The AUC values were compared using the DeLong test.Results The AUC values of the LR, RF, and GBM models in predicting the invasion depth of endometrial carcinoma were 0.780 (95% CI: 0.762 to 0.804), 0.860 (95% CI: 0.846 to 0.879), and 0.860 (95% CI: 0.843 to 0.877), respectively. The AUC values of the RF and GBM were the highest and equal. The DeLong test showed that there was a statistically significant difference in AUC values between LR, RF, and GBM models (P = 0.017, 0.023), while there was no statistically significant difference in AUC values between RF and GBM models (P = 3.310). The calibration curve and DCA curve show that all three models have good fit and clinical practicality.Conclusions The radiomics models based on ADC map have good value in predicting the invasion depth of EC.
[Keywords] endometrial carcinoma；myometrial invasion；magnetic resonance imaging；radiomics；machine learning；apparent diffusion coefficient

CUI Jing   GUO Ran   XIN Ruiqiang*  

Department of Radiology, Beijing Luhe Hospital, Captital Medical University, Beijing 101199, China

Corresponding author: XIN R Q, E-mail: rxin@ccmu.edu.cn

Conflicts of interest   None.

Received  2024-09-04
Accepted  2025-03-04
DOI: 10.12015/issn.1674-8034.2025.03.012
Cite this article as: CUI J, GUO R, XIN R Q. Predictive value of machine learning model based on ADC radiomics in evaluating the invasion depth of endometrial carcinoma[J]. Chin J Magn Reson Imaging, 2025, 16(3): 77-82. DOI:10.12015/issn.1674-8034.2025.03.012.

[1]
MAKKER V, MACKAY H, RAY-COQUARD I, et al. Endometrial cancer[J/OL]. Nat Rev Dis Primers, 2021, 7: 88 [2025-02-21]. https://pubmed.ncbi.nlm.nih.gov/34887451/. DOI: 10.1038/s41572-021-00324-8.
[2]
SIEGEL R L, MILLER K D, WAGLE N S, et al. Cancer statistics, 2023[J]. CA A Cancer J Clin, 2023, 73(1): 17-48. DOI: 10.3322/caac.21763.
[3]
PU C L, Biyuan, XU K, et al. Glycosylation and its research progress in endometrial cancer[J]. Clin Transl Oncol, 2022, 24(10): 1865-1880. DOI: 10.1007/s12094-022-02858-z.
[4]
ZHOU Q, WU X H, LIU J H, et al. Guidelines to the diagnosis and treatment of endometrial carcinoma(4th edition)[J]. Chin J Pract Gynecol Obstet, 2018, 34(8): 880-886. DOI: 10.19538/j.fk2018080114.
[5]
Gynecological Oncology Professional Committee of China Anti Cancer Association. Guidelines for diagnosis and treatment of endometrial cancer (2021 edition)[J]. China Oncol, 2021, 31(6): 501-512. DOI: 10.19401/j.cnki.1007-3639.2021.06.08.
[6]
ZHANG Q, YU X D, LIN M, et al. Multi-b-value diffusion weighted imaging for preoperative evaluation of risk stratification in early-stage endometrial cancer[J/OL]. Eur J Radiol, 2019, 119: 108637 [2025-02-21]. https://pubmed.ncbi.nlm.nih.gov/31446209/. DOI: 10.1016/j.ejrad.2019.08.006.
[7]
LARSON D M, CONNOR G P, BROSTE S K, et al. Prognostic significance of gross myometrial invasion with endometrial cancer[J]. Obstet Gynecol, 1996, 88(3): 394-398. DOI: 10.1016/0029-7844(96)00161-5.
[8]
STANZIONE A, CUOCOLO R, DEL GROSSO R, et al. Deep myometrial infiltration of endometrial cancer on MRI: a radiomics-powered machine learning pilot study[J]. Acad Radiol, 2021, 28(5): 737-744. DOI: 10.1016/j.acra.2020.02.028.
[9]
WANG L J, TSENG Y J, WEE N K, et al. Diffusion-weighted imaging versus dynamic contrast-enhanced imaging for pre-operative diagnosis of deep myometrial invasion in endometrial cancer: a meta-analysis[J]. Clin Imaging, 2021, 80: 36-42. DOI: 10.1016/j.clinimag.2021.06.027.
[10]
LIN G, HUANG Y T, CHAO A, et al. Endometrial cancer with cervical stromal invasion: diagnostic accuracy of diffusion-weighted and dynamic contrast enhanced MR imaging at 3T[J]. Eur Radiol, 2017, 27(5): 1867-1876. DOI: 10.1007/s00330-016-4583-0.
[11]
MAHESHWARI E, NOUGARET S, STEIN E B, et al. Update on MRI in evaluation and treatment of endometrial cancer[J]. Radiographics, 2022, 42(7): 2112-2130. DOI: 10.1148/rg.220070.
[12]
DE MUZIO F, FUSCO R, SIMONETTI I, et al. Functional assessment in endometrial and cervical cancer: diffusion and perfusion, two captivating tools for radiologists[J]. Eur Rev Med Pharmacol Sci, 2023, 27(16): 7793-7810. DOI: 10.26355/eurrev_202308_33435.
[13]
ZHANG G W, XU Z L, ZHENG J Y, et al. Prognostic value of multi b-value DWI in patients with locally advanced rectal cancer[J]. Eur Radiol, 2023, 33(3): 1928-1937. DOI: 10.1007/s00330-022-09159-7.
[14]
NOUGARET S, REINHOLD C, ALSHARIF S S, et al. Endometrial cancer: combined MR volumetry and diffusion-weighted imaging for assessment of myometrial and lymphovascular invasion and tumor grade[J]. Radiology, 2015, 276(3): 797-808. DOI: 10.1148/radiol.15141212.
[15]
CHEN J Y, FAN W M, GU H L, et al. The value of the apparent diffusion coefficient in differentiating type II from type I endometrial carcinoma[J]. Acta Radiol, 2021, 62(7): 959-965. DOI: 10.1177/0284185120944913.
[16]
ZHANG K Y, ZHANG Y, FANG X, et al. Nomograms of combining apparent diffusion coefficient value and radiomics for preoperative risk evaluation in endometrial carcinoma[J/OL]. Front Oncol, 2021, 11: 705456 [2025-02-21]. https://pubmed.ncbi.nlm.nih.gov/34386425/. DOI: 10.3389/fonc.2021.705456.
[17]
RAJA S, SHARMA P K, SUBRAMONIAN S G, et al. Enhancing preoperative assessment of endometrial cancer: the role of diffusion-weighted magnetic resonance imaging in evaluating myometrial invasion[J/OL]. Cureus, 2024, 16(6): e62111 [2025-02-21]. https://pubmed.ncbi.nlm.nih.gov/38993436/. DOI: 10.7759/cureus.62111.
[18]
GILLIES R J, KINAHAN P E, HRICAK H. Radiomics: images are more than pictures, they are data[J]. Radiology, 2016, 278(2): 563-577. DOI: 10.1148/radiol.2015151169.
[19]
MAYERHOEFER M E, MATERKA A, LANGS G, et al. Introduction to radiomics[J]. J Nucl Med, 2020, 61(4): 488-495. DOI: 10.2967/jnumed.118.222893.
[20]
DING S X, MENG H, YIN X P. Research progress of radiomics in endometrial cancer[J]. Chin J Magn Reson Imag, 2023, 14(4): 188-192. DOI: 10.12015/issn.1674-8034.2023.04.033.
[21]
LEFEBVRE T L, UENO Y, DOHAN A, et al. Development and validation of multiparametric MRI-based radiomics models for preoperative risk stratification of endometrial cancer[J]. Radiology, 2022, 305(2): 375-386. DOI: 10.1148/radiol.212873.
[22]
LIU J J. Study on the application of imageology in histological grading and muscle infiltration prediction of endometrial cancer[D]. Bangbu: Bengbu Medical College, 2023. DOI: 10.26925/d.cnki.gbbyc.2023.000419.
[23]
HAN Y Q. Predictive value of MRI features of whole uterus on myometrial infiltration depth of endometrial carcinoma[D]. Jinan: Shandong University, 2021. DOI: 10.27272/d.cnki.gshdu.2021.004414.
[24]
PRAKASAN A M, DHAS M, JAGATHNATHKRISHNA K M, et al. Prognostic factors for survival in patients with carcinoma endometrium[J]. South Asian J Cancer, 2023, 11(4): 309-314. DOI: 10.1055/s-0041-1735563.
[25]
MA Y, HOU M Y, ZHOU F, et al. Amide proton transfer imaging and intravoxel incoherent motion in evaluating risk stratification of early-stage endometrial cancer[J]. Chin J Med Imag, 2022, 30(6): 600-605. DOI: 10.3969/j.issn.1005-5185.2022.06.015.
[26]
ZHANG Y R, LI D D. Value of color ultrasound and CT in differentiating the pathological stage and the depth of myometrial invasion of endometrial cancer patients[J]. Chin J CT MRI, 2022, 20(9): 135-137. DOI: 10.3969/j.issn.1672-5131.2022.09.050.
[27]
YAN B, LIANG X F, ZHAO T T, et al. Preoperative prediction of deep myometrial invasion and tumor grade for stage I endometrioid adenocarcinoma: a simple method of measurement on DWI[J]. Eur Radiol, 2019, 29(2): 838-848. DOI: 10.1007/s00330-018-5653-2.
[28]
BERA K, BRAMAN N, GUPTA A, et al. Predicting cancer outcomes with radiomics and artificial intelligence in radiology[J]. Nat Rev Clin Oncol, 2022, 19(2): 132-146. DOI: 10.1038/s41571-021-00560-7.
[29]
NISHIKAWA T, OHKA F, AOKI K, et al. Easy-to-use machine learning system for the prediction of IDH mutation and 1p/19q codeletion using MRI images of adult-type diffuse gliomas[J]. Brain Tumor Pathol, 2023, 40(2): 85-92. DOI: 10.1007/s10014-023-00459-4.
[30]
SCAPICCHIO C, GABELLONI M, BARUCCI A, et al. A deep look into radiomics[J]. Radiol Med, 2021, 126(10): 1296-1311. DOI: 10.1007/s11547-021-01389-x.
[31]
UENO Y, FORGHANI B, FORGHANI R, et al. Endometrial carcinoma: MR imaging-based texture model for preoperative risk stratification-A preliminary analysis[J]. Radiology, 2017, 284(3): 748-757. DOI: 10.1148/radiol.2017161950.
[32]
YTRE-HAUGE S, DYBVIK J A, LUNDERVOLD A, et al. Preoperative tumor texture analysis on MRI predicts high-risk disease and reduced survival in endometrial cancer[J]. J Magn Reson Imaging, 2018, 48(6): 1637-1647. DOI: 10.1002/jmri.26184.
[33]
WANG X, SONG G, JIANG H T, et al. Can texture analysis based on single unenhanced CT accurately predict the WHO/ISUP grading of localized clear cell renal cell carcinoma?[J]. Abdom Radiol, 2021, 46(9): 4289-4300. DOI: 10.1007/s00261-021-03090-z.
[34]
KIM Y J. Machine learning models for sarcopenia identification based on radiomic features of muscles in computed tomography[J/OL]. Int J Environ Res Public Health, 2021, 18(16): 8710 [2025-02-21]. https://pubmed.ncbi.nlm.nih.gov/34444459/. DOI: 10.3390/ijerph18168710.
[35]
GUO R, SHEN X Z, XIN R Q, et al. Predictive value of random forest based on MRI radiomics in evaluating the invasion depth of endometrial carcinoma[J]. Chin J Magn Reson Imag, 2020, 11(12): 1133-1137. DOI: 10.12015/issn.1674-8034.2020.12.011.

PREV Prediction of lymphovascular space invasion in endometrial carcinoma based on preoperative multiparameter MRI deep transfer learning features
NEXT The value of T2WI imaging-based histology in the ability to identify penetrating placenta implantation
  


LINK


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