文献速递乳腺MR

MRI上表现为非肿块病变的恶性肿瘤与特发性肉芽肿性乳腺炎的鉴别:影像学、临床、影像组学及临床-影像组学模型

Differentiation of Malignancy and Idiopathic Granulomatous Mastitis Presenting as Non-mass Lesions on MRI: Radiological, Clinical, Radiomics, and Clinical-Radiomics Models.
PubMed 文献 · Acad Radiol · 2024-04-18

本研究纳入两中心2018年3月至2022年4月间178例乳腺MRI表现为非肿块强化(NME)的患者(特发性肉芽肿性乳腺炎69例、乳腺癌109例),比较影像学评估与基于Light GBM的临床、影像组学及临床-影像组学机器学习模型对两者的鉴别效能。结果显示影像科医师评估的AUC最低(训练集0.740、测试集0.737),临床-影像组学联合模型AUC最高(训练集0.979、测试集0.942)。作者认为该模型可能有助于准确鉴别表现为NME的特发性肉芽肿性乳腺炎与恶性病变。

要点

  • 特发性肉芽肿性乳腺炎组年龄更轻(p=0.014),脓肿、瘘管、皮肤改变均显著更常见(均p<0.001)。
  • 影像科医师评估NME的AUC为训练集0.740、测试集0.737,为各方法中最低。
  • 临床-影像组学联合模型AUC为训练集0.979、测试集0.942,为各方法中最高。

核心结论

  • 基于机器学习的临床-影像组学模型在评估NME区域时,可能具有准确鉴别特发性肉芽肿性乳腺炎与恶性病变的潜力。

研究设计与样本

  • 两中心研究,共纳入178例(特发性肉芽肿性乳腺炎69例、乳腺癌109例)2018年3月至2022年4月间乳腺MRI表现为NME的患者,数据分为训练集与外部测试集,采用Light GBM构建机器学习模型,参考标准为临床/病理诊断,并用CLEAR与METRICS工具评价报告质量与方法学质量。

关键影像表现/征象

  • 特发性肉芽肿性乳腺炎组年龄更轻(p=0.014),脓肿(p<0.001)、瘘管(p<0.001)、皮肤改变(p<0.001)显著更常见。
  • 两组病变大小无显著差异(p=0.213)。
  • 两名有经验的影像科医师依据乳腺影像报告和数据系统2013版词典评估MRI图像。
  • 病灶分别在T2加权、表观扩散系数及增强后T1加权序列上独立分割。

诊断标准与关键数值

  • 影像科医师评估NME的AUC为训练集0.740、测试集0.737,为各方法中最低。
  • 临床-影像组学联合模型AUC为训练集0.979、测试集0.942,为各方法中最高。
  • 研究构建并比较了影像学、临床、影像组学及临床-影像组学四种模型,并进行了决策曲线分析。

临床应用与阅片要点

  • 在MRI上评估表现为NME的病灶时,除影像学特征外,可结合年龄、皮肤改变、瘘管及脓肿等临床信息辅助鉴别。
  • 作者认为基于机器学习的临床-影像组学模型可能具有准确鉴别特发性肉芽肿性乳腺炎与恶性病变的潜力。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

RATIONALE AND OBJECTIVES: To investigate the effectiveness of machine learning-based clinical, radiomics, and combined models in differentiating idiopathic granulomatous mastitis (IGM) from malignancy, both presenting as non-mass enhancement (NME) lesions on magnetic resonance imaging (MRI), and to compare these models with radiological evaluation.

MATERIAL AND METHODS: A total of 178 patients (69 IGM and 109 breast cancer patients) with NME on breast MRI evaluated between March 2018 and April 2022, were included in this two-center study. Age, skin changes, presence of fistula, and abscess were recorded from hospital records. Two experienced radiologists evaluated MRI images according to the breast imaging reporting and data system 2013 lexicon. Lesions were segmented independently on T2-weighted, apparent diffusion coefficient, and post-contrast-T1-weighted sequences. Data were split into training and external testing sets. Machine learning models were built using Light GBM (light gradient-boosting machine). Radiological, clinical, radiomics, and clinical-radiomics models were created and compared. Decision curve analysis was performed. Quality of reporting and that of methodology were evaluated using CLEAR and METRICS tools.

RESULTS: IGM group was younger (p = 0.014). Abscesses (p < 0.001), fistulas (p < 0.001), and skin changes (p < 0.001) were significantly more common in the IGM group. No significant difference was detected in terms of lesion size (p = 0.213). In the evaluation of NME, the lowest performance belonged to the radiologists' evaluation (AUC for training, 0.740; for testing, 0.737), while the highest AUC was achieved by the model developed by combined clinical and radiomics features (AUC for training, 0.979; for testing, 0.942).

CONCLUSION: Our study has shown that the machine learning-based clinical-radiomics model might have the potential to accurately discriminate IGM and malignant lesions in evaluating NME areas.

来源:Acad Radiol,查看原文。版权归原作者所有。

引用

  1. Kayadibi Y, Saracoglu MS, Kurt SA, et al. Differentiation of Malignancy and Idiopathic Granulomatous Mastitis Presenting as Non-mass Lesions on MRI: Radiological, Clinical, Radiomics, and Clinical-Radiomics Models. Acad Radiol. 2024. PMID: 38641449. doi:10.1016/j.acra.2024.03.025

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。