Topology-based radiomic features for prediction of parotid gland cancer malignancy grade in magnetic resonance images
Topology-based radiomic features for prediction of parotid gland cancer malignancy grade in magnetic resonance images
复制标题
DOI：
10.1007/s10334-023-01084-0
复制
发表时间：
2023-04-20
影响因子：
2.3
通讯作者：
Ninomiya，Kenta
中科院分区：
文献类型：
--
作者：
Ikushima，Kojiro;Arimura，Hidetaka;Ninomiya，Kenta
目的评估腮腺癌 (PGC) 的恶性程度，以决定治疗政策。因此，我们研究了基于拓扑的放射组学特征在磁共振（MR）图像中预测腮腺癌（PGC）恶性程度的可行性。材料和方法本研究选择了39例PGC患者的二维T1和T2加权MR图像。 PGC 的成像特性可以使用拓扑进行量化，这对于使用 Betti 数不变量评估 PGC 区域中 k 维孔的数量或异质性非常有用。放射组学特征是根据使用弹性网络模型协调后获得的 41,472 个特征构建的。使用逻辑分类将 PGC 患者分为低/中和高级别恶性肿瘤组。使用合成少数过采样技术将训练数据增加四倍以避免过拟合问题。使用4倍交叉验证测试对所提出的方法进行评估。结果对于验证案例，所提出的方法的最高准确度为0.975，而传统方法的最高准确度为0.694。结论本研究表明，基于拓扑的放射组学特征对于PGC恶性程度的无创预测是可行的。
PurposeThe malignancy grades of parotid gland cancer (PGC) have been assessed for a decision of treatment policies. Therefore, we have investigated the feasibility of topology-based radiomic features for the prediction of parotid gland cancer (PGC) malignancy grade in magnetic resonance (MR) images.Materials and methodsTwo-dimensional T1- and T2-weighted MR images of 39 patients with PGC were selected for this study. Imaging properties of PGC can be quantified using the topology, which could be useful for assessing the number of the k-dimensional holes or heterogeneity in PGC regions using invariants of the Betti numbers. Radiomic signatures were constructed from 41,472 features obtained after a harmonization using an elastic net model. PGC patients were stratified using a logistic classification into low/intermediate- and high-grade malignancy groups. The training data were increased by four times to avoid the overfitting problem using a synthetic minority oversampling technique. The proposed approach was assessed using a 4-fold cross-validation test.ResultsThe highest accuracy of the proposed approach was 0.975 for the validation cases, whereas that of the conventional approach was 0.694.ConclusionThis study indicated that topology-based radiomic features could be feasible for the noninvasive prediction of the malignancy grade of PGCs.