Share：
Share this content in WeChat
X
[Chinese][PDF]341
• Clinical Article •
Identification of type Luminal and non-type Luminal breast cancers based on multiparametric MR habitat imaging
CHENG Weiqun  QI Xuan  YANG Hongkai  DUAN Shaofeng  HE Yongsheng  TONG Jinying  PAN Shuya  LIU Guangzhu 

Cite this article as: CHENG W Q, QI X, YANG H K, et al. Identification of type Luminal and non-type Luminal breast cancers based on multiparametric MR habitat imaging[J]. Chin J Magn Reson Imaging, 2025, 16(5): 170-180. DOI:10.12015/issn.1674-8034.2025.05.026.


[Abstract] Objective To explore the value of diagnosing Luminal and non-Luminal breast cancer (BC) based on multiparameter MR habitat imaging analysis.Materials and Methods A retrospective analysis was conducted on 216 cases of breast diseases treated at Ma'anshan People's Hospital from December 2019 to May 2024. These cases were confirmed by puncture biopsy or surgical pathology, including 147 cases of Luminal-type BC and 69 cases of non-Luminal-type BC. The patients' ages ranged from 26 to 85 years old, with an average age of (54.8 ± 10.9) years. The 216 patients were randomly divided into a training set and a validation set at a ratio of 7∶3. All patients underwent multi-parameter magnetic resonance imaging (mpMRI) scans. Image preprocessing was performed on the T2WI sequence, small field diffusion weighted imaging (ZOOMit-DWI) sequence in mpMRI, as well as the PEI, TTP, WASHIN, and WASHOUT sequences obtained from the analysis of the dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) sequence. Radiomics features were extracted from each functional parameter map. Cluster analysis was carried out through the Gaussian mixture model (GMM), and the clustering results were evaluated by the Silhouette coefficient. Finally, six groups of habitat images were generated, with each group containing three sub - regions. After image preprocessing, 1197 and 3591 omics features were extracted from the original and habitat sub-region images respectively. Redundant features were removed using robust normalization, Z-score standardization, min-max normalization, F test, the least absolute shrinkage and selection operator (LASSO) algorithm, and 10-fold cross-validation. Four features were selected from the clinical data to construct a clinical model. In the radiomics part, 4, 5, 10, 6, 11, and 9 features were selected, and six groups of radiomics models were established. Then, logistic regression was used to screen the radiomics models and clinical features to establish a combined model. In the habitat radiomics part, 14, 13, 19, 4, 14, and 13 features were selected, and six groups of habitat radiomics models were established. Logistic regression was used to screen the habitat radiomics models, intratumor heterogeneity score (ITH-score), and clinical features to establish a combined mode. The thresholds, sensitivities, specificities, accuracies, negative predictive values, and positive predictive values of the clinical model, radiomics model, habitat radiomics model, and their respective combined models were calculated respectively. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was calculated to analyze the diagnostic efficacy of each model. The DeLong test was used to compare the differences between groups pairwise, and decision curve analysis (DCA) was further used to evaluate the net benefit of the models.Results In the training set and validation set, there were statistically significant differences in ER, PR, Ki-67 and WHO grade between Luminal and non-Luminal BC (P < 0.05). In both the radiomics and habitat radiomics models, the Combine model had the best prediction performance. The AUC values of the Combine model in the training set and validation set were 0.967 and 0.798 respectively, and the optimal model was the MLP model among them. The AUC values of the Combine model in the training set and validation set were 0.969 and 0.910 respectively, and the optimal model was the linear_SVM model among them. Comparatively, the performance of the linear_SVM model was significantly better than that of the MLP model.Conclusions Analysis based on multi-parameter MR habitat imaging can diagnose Luminal-type and non-Luminal-type BC relatively accurately, which is helpful for the clinical diagnosis, treatment and management of BC.
[Keywords] breast cancer；magnetic resonance imaging；habitat imaging；radiomics；cluster analysis

CHENG Weiqun1, 2, 3   QI Xuan2, 3   YANG Hongkai2, 3   DUAN Shaofeng2, 3   HE Yongsheng2, 3*   TONG Jinying2, 3   PAN Shuya2, 3   LIU Guangzhu1, 2, 3  

1 Department of Radiology, Ma'anshan People's Hospital, Anhui Medical University, Ma'anshan 243000, China

2 Department of Radiology, Ma'anshan People's Hospital, Ma'anshan 243000, China

3 Ma'anshan Key Laboratory for Medical Image Modeling and Intelligent Analysis, Ma'anshan 243099, China

Corresponding author: HE Y S, E-mail: heyongsheng881@163.com

Conflicts of interest   None.

Received  2025-01-26
Accepted  2025-05-09
DOI: 10.12015/issn.1674-8034.2025.05.026
Cite this article as: CHENG W Q, QI X, YANG H K, et al. Identification of type Luminal and non-type Luminal breast cancers based on multiparametric MR habitat imaging[J]. Chin J Magn Reson Imaging, 2025, 16(5): 170-180. DOI:10.12015/issn.1674-8034.2025.05.026.

[1]
SIEGEL R L, GIAQUINTO A N, JEMAL A. Cancer statistics, 2024[J]. CA A Cancer J Clin, 2024, 74(1): 12-49. DOI: 10.3322/caac.21820.
[2]
ARNOLD M, MORGAN E, RUMGAY H, et al. Current and future burden of breast cancer: Global statistics for 2020 and 2040[J/OL]. Breast, 2022, 66: 15-23 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/36084384/. DOI: 10.1016/j.breast.2022.08.010.
[3]
MA M W, LIU R Y, WEN C J, et al. Predicting the molecular subtype of breast cancer and identifying interpretable imaging features using machine learning algorithms[J]. Eur Radiol, 2022, 32(3): 1652-1662. DOI: 10.1007/s00330-021-08271-4.
[4]
GENG T, WU Z Q, QIN G G, et al. Research progress of radiomics and radiogenomics in predicting the molecular subtypes of breast cancer[J]. Int J Med Radiol, 2024, 47(3): 306-309. DOI: 10.19300/j.2024.Z21290.
[5]
HE Y S, DUAN S F, WANG W L, et al. Integrative radiomics clustering analysis to decipher breast cancer heterogeneity and prognostic indicators through multiparametric MRI[J/OL]. NPJ Breast Cancer, 2024, 10(1): 72 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/39112498/. DOI: 10.1038/s41523-024-00678-8.
[6]
ZHU J J, GENG J H, SHAN W, et al. Development and validation of a deep learning model for breast lesion segmentation and characterization in multiparametric MRI[J/OL]. Front Oncol, 2022, 12: 946580 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/36033449/. DOI: 10.3389/fonc.2022.946580.
[7]
LAI S S, LIANG F R, ZHANG W L, et al. Evaluation of molecular receptors status in breast cancer using an mpMRI-based feature fusion radiomics model: mimicking radiologists' diagnosis[J/OL]. Front Oncol, 2023, 13: 1219071 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/38074664/. DOI: 10.3389/fonc.2023.1219071.
[8]
WANG G S, GUO Q, SHI D F, et al. Clinical breast MRI-based radiomics for distinguishing benign and malignant lesions: an analysis of sequences and enhanced phases[J]. J Magn Reson Imaging, 2024, 60(3): 1178-1189. DOI: 10.1002/jmri.29150.
[9]
SHI Z W, LIU Z Y. The challenges and solutions in radiomics study[J]. Chin J Radiol, 2022, 56(1): 9-11. DOI: 10.3760/cma.j.cn112149-20211111-00998.
[10]
VAN TIMMEREN J E, CESTER D, TANADINI-LANG S, et al. Radiomics in medical imaging- "how-to" guide and critical reflection[J/OL]. Insights Imaging, 2020, 11(1): 91 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/32785796/. DOI: 10.1186/s13244-020-00887-2.
[11]
LIU Y H, GAO Y. Implications of habitat imaging-based multisequence MRI in adult-type diffuse glioma[J]. Chin J Magn Reson Imag, 2023, 14(9): 119-124. DOI: 10.12015/issn.1674-8034.2023.09.022.
[12]
LI S L, DAI Y M, CHEN J Y, et al. MRI-based habitat imaging in cancer treatment: current technology, applications, and challenges[J/OL]. Cancer Imaging, 2024, 24(1): 107 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/39148139/. DOI: 10.1186/s40644-024-00758-9.
[13]
LIU J C, WU H, FANG J Q, et al. State-of-art of habitat imaging of glioblastoma based on multi-parameter MRI[J]. Int J Med Radiol, 2021, 44(4): 420-424. DOI: 10.19300/j.2021.Z18816.
[14]
JIANG L, YOU C, XIAO Y, et al. Radiogenomic analysis reveals tumor heterogeneity of triple-negative breast cancer[J/OL]. Cell Rep Med, 2022, 3(7): 100694 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/35858585/. DOI: 10.1016/j.xcrm.2022.100694.
[15]
UYSAL E, TOPALOĞLU Ö F, ARı A, et al. Can magnetic resonance imaging texture analysis change the breast imaging reporting and data system category of breast lesions?[J/OL]. Clin Imaging, 2023, 97: 44-49 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/36889114/. DOI: 10.1016/j.clinimag.2023.02.016.
[16]
QIAN L, LI T, TANG C Y. The diagnostic value of gray level co-occurrence matrix texture features based on different ROIs of TIRM sequence in breast masses[J]. Chin J CT MRI, 2023, 21(9): 101-103. DOI: 10.3969/j.issn.1672-5131.2023.09.034.
[17]
WANG M, WANG X L, ZHANG J, et al. The value of identifying benign and malignant breast lesions based on the texture feature of Tirm sequence combined with time-intensity curve[J]. Chin J Magn Reson Imag, 2021, 12(6): 83-87. DOI: 10.12015/issn.1674-8034.2021.06.016.
[18]
ZHANG J H, WANG X L, ZHANG L X, et al. Radiomics predict postoperative survival of patients with primary liver cancer with different pathological types[J/OL]. Ann Transl Med, 2020, 8(13): 820 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/32793665/. DOI: 10.21037/atm-19-4668.
[19]
LI Y J, ZENG X R, LIN C W, et al. Simultaneous estimation of cluster number and feature sparsity in high-dimensional cluster analysis[J]. Biometrics, 2022, 78(2): 574-585. DOI: 10.1111/biom.13449.
[20]
LI J Q, QIU Z B, ZHANG C, et al. ITHscore: comprehensive quantification of intra-tumor heterogeneity in NSCLC by multi-scale radiomic features[J]. Eur Radiol, 2023, 33(2): 893-903. DOI: 10.1007/s00330-022-09055-0.
[21]
WANG X, BAI H, ZHANG J Y, et al. Genetic intratumor heterogeneity remodels the immune microenvironment and induces immune evasion in brain metastasis of lung cancer[J]. J Thorac Oncol, 2024, 19(2): 252-272. DOI: 10.1016/j.jtho.2023.09.276.
[22]
RANJBARZADEH R, CAPUTO A, TIRKOLAEE E B, et al. Brain tumor segmentation of MRI images: a comprehensive review on the application of artificial intelligence tools[J/OL]. Comput Biol Med, 2023, 152: 106405 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/36512875/. DOI: 10.1016/j.compbiomed.2022.106405.
[23]
JEONG S Y, PARK J E, KIM N, et al. Hypovascular cellular tumor in primary central nervous system lymphoma is associated with treatment resistance: tumor habitat analysis using physiologic MRI[J]. AJNR Am J Neuroradiol, 2022, 43(1): 40-47. DOI: 10.3174/ajnr.A7351.
[24]
BAILO M, PECCO N, CALLEA M, et al. Decoding the heterogeneity of malignant gliomas by PET and MRI for spatial habitat analysis of hypoxia, perfusion, and diffusion imaging: a preliminary study[J/OL]. Front Neurosci, 2022, 16: 885291 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/35911979/. DOI: 10.3389/fnins.2022.885291.
[25]
DONG F F, CHEN J, LIU F Y, et al. Modeling and prediction of set-up errors in breast cancer image-guided radiotherapy using the Gaussian mixture model[J/OL]. Oncol Lett, 2024, 28(6): 573 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/39397807/. DOI: 10.3892/ol.2024.14706.
[26]
ELOYAN A, YUE M S, KHACHATRYAN D. Tumor heterogeneity estimation for radiomics in cancer[J]. Stat Med, 2020, 39(30): 4704-4723. DOI: 10.1002/sim.8749.
[27]
JUAN-ALBARRACÍN J, FUSTER-GARCIA E, GARCÍA-FERRANDO G A, et al. ONCOhabitats: a system for glioblastoma heterogeneity assessment through MRI[J/OL]. Int J Med Inform, 2019, 128: 53-61 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/31160012/. DOI: 10.1016/j.ijmedinf.2019.05.002.
[28]
CHANG Y C, ACKERSTAFF E, TSCHUDI Y, et al. Delineation of tumor habitats based on dynamic contrast enhanced MRI[J/OL]. Sci Rep, 2017, 7(1): 9746 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/28851989/. DOI: 10.1038/s41598-017-09932-5.
[29]
NAPEL S, MU W, JARDIM-PERASSI B V, et al. Quantitative imaging of cancer in the postgenomic era: Radio(geno)mics, deep learning, and habitats[J]. Cancer, 2018, 124(24): 4633-4649. DOI: 10.1002/cncr.31630.
[30]
GAUSTAD J V, HAUGE A, WEGNER C S, et al. DCE-MRI of tumor hypoxia and hypoxia-associated aggressiveness[J/OL]. Cancers (Basel), 2020, 12(7): 1979 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/32698525/. DOI: 10.3390/cancers12071979.
[31]
WINFIELD J M, PAYNE G S, WELLER A, et al. DCE-MRI, DW-MRI, and MRS in cancer: challenges and advantages of implementing qualitative and quantitative multi-parametric imaging in the clinic[J]. Top Magn Reson Imaging, 2016, 25(5): 245-254. DOI: 10.1097/RMR.0000000000000103.
[32]
XU R, YU D, LUO P, et al. Do habitat MRI and fractal analysis help distinguish triple-negative breast cancer from non-triple-negative breast carcinoma[J]. Can Assoc Radiol J, 2024, 75(3): 584-592. DOI: 10.1177/08465371241231573.
[33]
CHEN H Q, LIU Y L, ZHAO J Q, et al. Quantification of intratumoral heterogeneity using habitat-based MRI radiomics to identify HER2-positive, -low and-zero breast cancers: a multicenter study[J/OL]. Breast Cancer Res, 2024, 26(1): 160 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/39578913/. DOI: 10.1186/s13058-024-01921-7.
[34]
HUANG Y N, YAO Z, LI L L, et al. Deep learning radiopathomics based on preoperative US images and biopsy whole slide images can distinguish between luminal and non-luminal tumors in early-stage breast cancers[J/OL]. EBioMedicine, 2023, 94: 104706 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/37478528/. DOI: 10.1016/j.ebiom.2023.104706.
[35]
HUANG T, FAN B, QIU Y Y, et al. Application of DCE-MRI radiomics signature analysis in differentiating molecular subtypes of luminal and non-luminal breast cancer[J/OL]. Front Med (Lausanne), 2023, 10: 1140514 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/37181350/. DOI: 10.3389/fmed.2023.1140514.
[36]
LIU L, MEI N, YIN B, et al. Correlation of DCE-MRI perfusion parameters and molecular biology of breast infiltrating ductal carcinoma[J/OL]. Front Oncol, 2021, 11: 561735 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/34722229/. DOI: 10.3389/fonc.2021.561735.
[37]
SONG S E, CHO K R, CHO Y, et al. Machine learning with multiparametric breast MRI for prediction of Ki-67 and histologic grade in early-stage luminal breast cancer[J]. Eur Radiol, 2022, 32(2): 853-863. DOI: 10.1007/s00330-021-08127-x.
[38]
KREIPE H, HARBECK N, CHRISTGEN M. Clinical validity and clinical utility of Ki67 in early breast cancer[J/OL]. Ther Adv Med Oncol, 2022, 14: 17588359221122725 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/36105888/. DOI: 10.1177/17588359221122725.
[39]
SMITH I, ROBERTSON J, KILBURN L, et al. Long-term outcome and prognostic value of Ki67 after perioperative endocrine therapy in postmenopausal women with hormone-sensitive early breast cancer (POETIC): an open-label, multicentre, parallel-group, randomised, phase 3 trial[J]. Lancet Oncol, 2020, 21(11): 1443-1454. DOI: 10.1016/S1470-2045(20)30458-7.
[40]
BENNANI-BAITI B, PINKER K, ZIMMERMANN M, et al. Non-invasive assessment of hypoxia and neovascularization with MRI for identification of aggressive breast cancer[J/OL]. Cancers (Basel), 2020, 12(8): 2024 [2025-01-25]. https://pubmed.ncbi.nlm.nih.gov/32721996/. DOI: 10.3390/cancers12082024.
[41]
QI X, WANG W L, PAN S Y, et al. Predictive value of triple negative breast cancer based on DCE-MRI multi-phase full-volume ROI clinical radiomics model[J]. Acta Radiol, 2024, 65(2): 173-184. DOI: 10.1177/02841851231215145.

PREV Interpretable machine learning model for predicting preoperative histological grade of invasive breast cancer based on high resolution delay period of magnetic resonance imaging
NEXT Combined multi-b-value DWI and DCE distributed parameter model in diagnosing radiation necrosis: One case report
  


LINK


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