| 摘要: |
| 目的 探讨光谱CT联合MRI在鉴别Luminal型与非Luminal型乳腺癌中的价值。 方法 回顾性收集2024年2月至2025年4月在浙江中医药大学附属第一医院接受诊治并经病理证实为乳腺癌的患者。所有患者在术前同时接受光谱CT胸部平扫和乳腺MRI增强检查，最终共入组110例。按照免疫组织化学染色结果，将所有患者分为Luminal 型与非Luminal 型两组。将光谱CT胸部平扫重建为多参数图像，包括有效原子序数图（Zeff）、电子云密度图（EDW）、虚拟单能级40 keV和100 keV图像，并计算光谱曲线斜率（K）。MRI乳腺增强图像记录时间信号强度（TIC）曲线类型、病灶边缘、病灶形态、肿瘤最大径和表观扩散系数（ADC）。根据数据的正态分布情况，采用独立样本t检验或Mann-Whitney U检验比较两组患者的光谱CT参数和MRI参数的差异。通过单因素logistic回归分析后差异有统计学意义的临床及影像学指标，继而进行多因素logistic回归分析，构建多模态联合模型。最终采用受试者工作特征（ROC）曲线评估模型的诊断效能。 结果 Luminal型和非Luminal型乳腺癌患者在年龄、月经史、肿瘤的最大径、ADC、40 keV时的CT值、100 keV时的CT值和EDW上的差异均具有统计学意义，P<0.05。其中，肿瘤的最大径、ADC值和EDW是Luminal型和非Luminal型乳腺癌的独立预测因素。ROC曲线显示，相比于单一模态，基于光谱CT和MRI联合的多模态模型的诊断效能最好（AUC=0.867，灵敏度85.42，特异度71.40）。 结论 利用光谱CT联合MRI构建的多模态影像学模型在鉴别Luminal型和非Luminal型乳腺癌中具有一定潜力，可以为临床诊疗提供参考依据。 |
| 关键词: 乳腺肿瘤 能谱成像 MRI Luminal型 非Luminal型 |
| DOI： |
| 分类号: |
| 基金项目:浙江省中医药科技计划项目，浙江省医药卫生科技计划项目 |
|
| The value of spectral CT combined with MRI in differentiating Luminal and non-Luminal breast cancer |
|
吴郑鑫1, Zhou Zhaoliang1, Yang Panpan1, Dong Shushan2, Wang Shiwei1, Zhang Ruixin3
|
|
1.Department of Radiology, The First Affiliated Hospital of Zhejiang Chinese Medical University (Zhejiang Provincial Hospital of Chinese Medicine);2.Philips Healthcare;3.The First Affiliated Hospital of Zhejiang Chinese Medical University (Zhejiang Provincial Hospital of Chinese Medicine)
|
| Abstract: |
| Objective Explore the value of spectral CT combined with MRI in differentiating Luminal and non-Luminal breast cancer. Methods Retrospectively collect the clinical, pathological and imaging data of breast cancer patients who received both Spectral CT chest plain scan and breast MRI enhanced scan before surgery in the First Affiliated Hospital of Zhejiang Chinese Medical University from November 2024 to April 2025. A total of 110 patients were included. Based on the immunohistochemistry results, these patients were divided into two groups: Luminal type and non-Luminal type. Spectral CT chest plain scan was used to reconstruct multi-parameter images, including effective atomic number (Zeff) map, electron cloud density (EDW) map, virtual single-level images at 40 keV and 100 keV, and the slope of spectral curve (K) was calculated. Time signal intensity (TIC) curve type, lesion edge, lesion shape, tumor maximum diameter and apparent diffusion coefficient (ADC) were recorded on the images breast enhanced MRI. Based on the normal distribution of the data, independent sample t-test or Mann-Whitney U test was used to compare the differences in spectral CT multi-parameters and MRI parameters between the two groups of patients. After univariate logistic regression analysis, the clinical and imaging indicators with statistically significant differences were further analyzed by multivariate logistic regression to construct a multimodal combined model. Finally, the diagnostic efficacy of the model was evaluated by the receiver operating characteristic (ROC) curve. Results Statistically significant differences were found in age, menstrual history, maximum diameter, ADC, CT value at 40 keV, CT value at 100 keV and EDW between Luminal and non-Luminal breast cancer patients. Moreover, the maximum diameter, ADC and EDW were proved to be the independent predictors for Luminal and non-Luminal breast cancer. The ROC curve showed that the combined model constructed based on these three independent predictors had good diagnostic efficacy (the area under the curve was 0.867, the sensitivity was 85.42%, and the specificity was 71.40%). Conclusion The combination of spectral CT and MRI has certain potential in differentiating Luminal and non-Luminal breast cancer, and can provide a reference basis for clinical diagnosis and treatment. |
| Key words: Breast tumor, Spectral imaging, MRI, Luminal type, Non-Luminal type |