Share：
Share this content in WeChat
X
[Chinese] [PDF] 176 7
• Clinical Article •
Differentiating non-mass breast cancer and non-lactational mastitis based on multi-parameter MRI radiomics
SONG Lijun  XUE Zhiwei  TIAN Xiongling  JIA Yi  MAYIDILI·Nijiati  

DOI:10.12015/issn.1674-8034.2025.08.011.


[Abstract] Objective To investigate the value of imaging omics model based on multimodal magnetic resonance imaging (MRI) in the differential diagnosis of non mass breast cancer and non lactating mastitis (NLM).Materials and Methods The MRI data of 193 patients with non mass breast cancer and NLM confirmed by pathology in the First Affiliated Hospital of Traditional Chinese Medicine, Xinjiang Medical University from June 2020 to June 2024 were retrospectively collected, including 100 cases of non mass breast cancer and 93 cases of NLM. The total number of lesions in the two groups was 225, including 110 breast cancer (48.89%) and 115 NLM (51.11%). It is randomly divided into training set (157 cases) and test set (68 cases) according to 7∶3. The support vector machines (SVM) learning algorithm was used to construct single sequence models and multi parameter MRI models for the first, fourth, and seventh phases of dynamic contrast-enhanced magnetic resonance imaging (CE1, CE4, CE7), T2 weighted imaging (T2WI) and diffusion weighted imaging (DWI). The fusion model was established by combining the data of five sequences and clinical characteristics. The performance of different models was evaluated by receiver operating characteristic (ROC) curve, calibration curve and decision curve analysis (DCA), and the model was interpreted and visualized using shap graphics.Results The area under the curve (AUC) of CE1, CE4, CE7, T2WI and DWI sequences in the test set were 0.768, 0.804, 0.746, 0.769 and 0.812, respectively. The AUC of DWI in the test set was the highest, followed by CE4; the AUC of the multi parameter MRI model in the test set was 0.840 (95% confidence interval was 0.749 to 0.932), while the AUC of the fusion model in the test set was 0.866 (95% confidence interval was 0.783 to 0.948), which was significantly different from CE1, CE4, CE7 and T2WI single-mode models (P < 0.01). The results showed that the accuracy of the integrated model was the highest (77.94%); the sensitivity of the integrated model was the highest (90.00%); and the specificity of the integrated model and the CE4 sequence was the highest (both at 68.42%).Conclusions The fusion model of multi parameter MRI combined with clinical features has higher accuracy, sensitivity and specificity, and better prediction performance than the single sequence model and multi-parametric MRI models, which can provide higher value for the differential diagnosis of non mass breast cancer and NLM.
[Keywords] non lactation mastitis；non mass breast cancer；multiparameter；magnetic resonance imaging；differential diagnosis

SONG Lijun1, 2   XUE Zhiwei2   TIAN Xiongling2   JIA Yi2   MAYIDILI·Nijiati 2*  

1 Xinjiang Key Laboratory of Artificial Intelligence Assisted Imaging Diagnosis, Kashi 844000, China

2 Imaging Center of Xinjiang Medical University Affiliated Traditional Chinese Medicine Hospital, Urumqi 830000, China

Corresponding author: NIJIATI M, E-mail: 1376906729@qq.com

Conflicts of interest   None.

Received  2025-03-11
Accepted  2025-07-31
DOI: 10.12015/issn.1674-8034.2025.08.011
DOI:10.12015/issn.1674-8034.2025.08.011.

[1]
HAN B F, ZHENG R S, ZENG H M, et al. Cancer incidence and mortality in China, 2022[J]. J Natl Cancer Cent, 2024, 4(1): 47-53. DOI: 10.1016/j.jncc.2024.01.006.
[2]
LUAN Y S, LI Y Y, ZHU A Y, et al. The value of color Doppler ultrasound in the diagnosis of non-puerperal mastitis[J]. Chin J Ultrasound Med, 2021, 37(1): 25-27. DOI: 10.3969/j.issn.1002-0101.2021.01.008.
[3]
WU X Y, ZHAN S H, LU M Y, et al. Value of intravoxel incoherent motion (IVIM) model in differential diagnosis of idiopathic granulomatous mastitis and invasive ductal carcinoma[J]. Chin Imag J Integr Tradit West Med, 2022, 20(6): 558-562. DOI: 10.3969/j.issn.1672-0512.2022.06.012.
[4]
ZHOU H Y, LI H, WANG S L, et al. Clinical analysis of misdiagnosis of granulomatous lobular mastitis[J]. Clin Misdiagnosis Mistherapy, 2024, 37(6): 10-13. DOI: 10.3969/j.issn.1002-3429.2024.06.003.
[5]
ZHAO Y J, CHEN R, ZHANG T, et al. MRI-based machine learning in differentiation between benign and malignant breast lesions[J/OL]. Front Oncol, 2021, 11: 552634 [2025-03-10]. https://pubmed.ncbi.nlm.nih.gov/34733774/. DOI: 10.3389/fonc.2021.552634.
[6]
LI Y, YANG Z L, LV W Z, et al. Non-mass enhancements on DCE-MRI: development and validation of a radiomics-based signature for breast cancer diagnoses[J/OL]. Front Oncol, 2021, 11: 738330 [2025-03-10]. https://pubmed.ncbi.nlm.nih.gov/34631572/. DOI: 10.3389/fonc.2021.738330.
[7]
CHANG L F, LAN H L. Effect of neoadjuvant chemotherapy on angiogenesis and cell proliferation of breast cancer evaluated by dynamic enhanced magnetic resonance imaging[J/OL]. Biomed Res Int, 2022, 2022: 3156093 [2025-03-10]. https://pubmed.ncbi.nlm.nih.gov/35915805/. DOI: 10.1155/2022/3156093.
[8]
GONG W Y, ZHENG Y Q, LI X R. Research advances in magnetic resonance imaging in non-lactating mastitis[J]. Acad J Chin PLA Med Sch, 2023, 44(10): 1167-1171. DOI: 10.12435/j.issn.2095-5227.2023.045.
[9]
LIU P, YU X J, LI C Z, et al. Differential diagnosis of dynamic contrast-enhanced-MRI-based radiomics model for granulomatous mastitis and breast cancer[J]. Chin J Med Imag, 2024, 32(2): 144-149. DOI: 10.3969/j.issn.1005-5185.2024.02.007.
[10]
RUAN J J, CHEN X S, SU J J, et al. The value of ultrasonography in differential diagnosis of plasma cell mastitis and invasive ductal carcinoma[J]. Chin J Ultrasound Med, 2022, 38(9): 984-987. DOI: 10.3969/j.issn.1002-0101.2022.09.007.
[11]
ZHENG Y, WANG L, HAN X, et al. Combining contrast-enhanced ultrasound and blood cell analysis to improve diagnostic accuracy of plasma cell mastitis[J]. Exp Biol Med (Maywood), 2022, 247(2): 97-105. DOI: 10.1177/15353702211049361.
[12]
LI F, XU M L, ZENG S E, et al. Differential diagnosis of massive granulomatous mastitis and invasive ductal carcinoma by histogram analysis of ultrasound gray[J]. Chin J Med Imag, 2020, 28(8): 602-606. DOI: 10.3969/j.issn.1005-5185.2020.08.011.
[13]
NGUYEN M H, MOLLAND J G, KENNEDY S, et al. Idiopathic granulomatous mastitis: case series and clinical review[J]. Intern Med J, 2021, 51(11): 1791-1797. DOI: 10.1111/imj.15112.
[14]
VELIDEDEOGLU M, UMMAN V, KILIC F, et al. Idiopathic granulomatous mastitis: introducing a diagnostic algorithm based on 5 years of follow-up of 152 cases from Turkey and a review of the literature[J]. Surg Today, 2022, 52(4): 668-680. DOI: 10.1007/s00595-021-02367-6.
[15]
LI C Z, WANG W S, REN H, et al. MR imaging comparative study of granulomatous mastitis and invasive breast cancer[J]. J Med Imag, 2023, 33(10): 1800-1803.
[16]
CHEN H Y, HAO X P, LI Q, et al. Clinical characteristics and misdiagnosed analysis of 25 patients with non-lactating mastitis[J]. Clin Misdiagnosis Mistherapy, 2022, 35(2): 12-15. DOI: 10.3969/j.issn.1002-3429.2022.02.004.
[17]
CAO X H, YANG L. Clinical analysis of plasma cell mastitis misdiagnosed as breast cancer and intraductal Papilloma before operation[J]. Clin Misdiagnosis Mistherapy, 2023, 36(7): 1-4. DOI: 10.3969/j.issn.1002-3429.2023.07.001.
[18]
GUAN C G, SHAO S, ZHENG N, et al. Application value of radiomics based on DCE-MRI combined with DKI in predicting triple-negative breast cancer[J]. Chin J Magn Reson Imag, 2025, 16(2): 35-43. DOI: 10.12015/issn.1674-8034.2025.02.006.
[19]
LIU B B, ZHANG Y, WANG J. Diagnostic value of DCE-MRI combined with MIP and breast mammography in non-mass enhancement lesions[J]. Chin Imag J Integr Tradit West Med, 2022, 20(5): 449-453. DOI: 10.3969/j.issn.1672-0512.2022.05.012.
[20]
GAO L, WANG X P, LI B. Analysis of DCE-MRI and DWI in the differential diagnosis of breast cancer and granulomatous mastitis[J]. Chinese Journal of CT and MRI, 2025, 23(5): 95-98. DOI: 10.3969/j.issn.1672-5131.2025.05.028.
[21]
ZHOU X, WANG Z, FU Y F. Analysis of DCE-MRI features and diagnostic value of non-mass-enhancing benign and malignant breast lesions[J]. Chin J CT MRI, 2024, 22(7): 118-120. DOI: 10.3969/j.issn.1672-5131.2024.07.037.
[22]
WANG W J, LI J J, WANG Z M, et al. Application of deep learning model fusion of T2WI and DCE-MRI in breast lesion classification[J]. J Clin Radiol, 2024, 43(11): 1871-1876. DOI: 10.13437/j.cnki.jcr.2024.11.033.
[23]
LV S Y, ZHAO Q F, WANG X Y, et al. The value of radiomics in discriminating granulomatous lobular mastitis from invasive breast cancer in non-mass-like enhancing lesions[J]. J Clin Radiol, 2025, 44(2): 253-258. DOI: 10.13437/j.cnki.jcr.2025.02.003.
[24]
WU L H, YANG W, ZHOU X P, et al. Clinical features, mammography and MRI manifestations for differentiating non-mass breast cancer and mastitis[J]. Chin J Med Imag Technol, 2023, 39(11): 1653-1658. DOI: 10.13929/j.issn.1003-3289.2023.11.013.
[25]
HE R H, LI L, QIAN W J. The differential diagnosis value of DCE-MRI, multi-b value DWI combined with CA153 and CEA in breast cancer and lumpy plasma cell mastitis[J]. Chin J CT MRI, 2023, 21(6): 89-92. DOI: 10.3969/j.issn.1672-5131.2023.06.030.
[26]
FENG Y H. Value of DWI and ADC combined enhancement in the differential diagnosis of non-mass breast cancer and inflammatory lesions[J]. World J Complex Med, 2023, 9(6): 38-41. DOI: 10.11966/j.issn.2095-994X.2023.09.06.11.
[27]
TONG Y, MI N, ZHANG R, et al. Evaluation of short-term efficacy of neoadjuvant chemotherapy and pathological changes in patients with advanced breast cancer by DWI[J]. Imag Sci Photochem, 2020, 38(2): 307-312. DOI: 10.7517/issn.1674-0475.190919.
[28]
LI L H, WANG F. The value of contrast-enhanced ultrasound and dynamic contrast-enhanced magnetic resonance imaging in the diagnosis of breast non-mass-like enhancement lesions[J]. Imag Sci Photochem, 2021, 39(3): 381-385. DOI: 10.7517/issn.1674-0475.201111.
[29]
XU W H, LIU J L, WU Y X. Value of multimodal magnetic resonance imaging based on artificial intelligence in differentiating benign and malignant breast lesions[J]. J Med Inf, 2024, 37(24): 144-147. DOI: 10.3969/j.issn.1006-1959.2024.24.039.

PREV Predictive value of electrocardiographic Q waves and CMR myocardial strain for microcirculatory obstruction after PCI treatment in patients with acute ST-elevation myocardial infarction
NEXT A study on the prediction of preoperative risk stratification of hepatocellular carcinoma based on multi-phase MRI radiomics combined with different machine learning models
  


LINK


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