Breast MRI radiomics and machine learning radiomics-based predictions of response to neoadjuvant chemotherapy – how are they affected by variations in tumour delineation?
paper_authors: Sepideh Hatamikia, Geevarghese George, Florian Schwarzhans, Amirreza Mahbod, Ramona Woitek for: 这个研究的目的是为了evaluating the impact of variations in manual delineations of volumes of interest (VOIs) on the performance of radiomics predictors in breast cancer subtypes.methods: 这个研究使用了contrast-enhanced magnetic resonance imaging (MRI) acquired prior to treatment (baseline MRI scans)，并使用了不同的mathematical operations such as erosion, smoothing, dilation, randomization, and ellipse fitting to simulate variations of segmentation masks.results: 研究发现，使用不同的VOI delineation methods can significantly affect the number of robust features and prediction performance in radiomics analysis. Specifically, smoothing and erosion yielded the highest number of robust features and the best prediction performance, while ellipse fitting and dilation led to the lowest robustness and prediction performance for both breast cancer subtypes. Additionally, the study found that at most 28% of the selected features were similar to manual VOIs when different VOI delineation data were used.Abstract
Manual delineation of volumes of interest (VOIs) by experts is considered the gold-standard method in radiomics analysis. However, it suffers from inter- and intra-operator variability. A quantitative assessment of the impact of variations in these delineations on the performance of the radiomics predictors is required to develop robust radiomics based prediction models. In this study, we developed radiomics models for the prediction of pathological complete response to neoadjuvant chemotherapy in patients with two different breast cancer subtypes based on contrast-enhanced magnetic resonance imaging acquired prior to treatment (baseline MRI scans). Different mathematical operations such as erosion, smoothing, dilation, randomization, and ellipse fitting were applied to the original VOIs delineated by experts to simulate variations of segmentation masks. The effects of such VOI modifications on various steps of the radiomics workflow, including feature extraction, feature selection, and prediction performance, were evaluated. Using manual tumor VOIs and radiomics features extracted from baseline MRI scans, an AUC of up to 0.96 and 0.89 was achieved for human epidermal growth receptor 2 positive and triple-negative breast cancer, respectively. For smoothing and erosion, VOIs yielded the highest number of robust features and the best prediction performance, while ellipse fitting and dilation lead to the lowest robustness and prediction performance for both breast cancer subtypes. At most 28% of the selected features were similar to manual VOIs when different VOI delineation data were used. Differences in VOI delineation affects different steps of radiomics analysis, and their quantification is therefore important for development of standardized radiomics research.
摘要
临床验证是验证分析的标准方法，但它受到Operator variability的影响。为了开发可靠的验证模型，我们需要评估随着分割masks的变化而导致的预测器性能的影响。我们在基线MRI扫描前进行了针对不同乳腺癌分型的预后化学治疗的预测，并使用不同的数学操作来模拟分割masks的变化。我们评估了这些变化对验证过程中的特征提取、特征选择和预测性能的影响。使用手动肿瘤分割和基线MRI扫描中提取的验证特征，我们可以达到0.96和0.89的AUC，分别为人类肿瘤抑制剂2阳性和三重阴性乳腺癌。对于平滑和减小操作，分割masks具有最高的Robust特征数和最好的预测性能，而 для�elia和扩大操作，分割masks具有最低的Robust特征数和预测性能。最多28%的选择特征与手动分割数据相同。不同的分割masks导致不同的预测过程中的不同步骤受到影响，因此其量化对于开发标准化验证研究非常重要。