Department of Radiation Oncology, UC San Francisco, San Francisco, CA 94143, United States of America.
Department of Radiation Oncology, UC Los Angeles, Los Angeles, CA 90095, United States of America.
Phys Med Biol. 2024 Jul 25;69(15). doi: 10.1088/1361-6560/ad64b8.
We aim to develop a Multi-modal Fusion and Feature Enhancement U-Net (MFFE U-Net) coupling with stem cell niche proximity estimation to improve voxel-wise Glioblastoma (GBM) recurrence prediction.57 patients with pre- and post-surgery magnetic resonance (MR) scans were retrospectively solicited from 4 databases. Post-surgery MR scans included two months before the clinical diagnosis of recurrence and the day of the radiologicaly confirmed recurrence. The recurrences were manually annotated on the T1ce. The high-risk recurrence region was first determined. Then, a sparse multi-modal feature fusion U-Net was developed. The 50 patients from 3 databases were divided into 70% training, 10% validation, and 20% testing. 7 patients from the 4th institution were used as external testing with transfer learning. Model performance was evaluated by recall, precision, F1-score, and Hausdorff Distance at the 95% percentile (HD95). The proposed MFFE U-Net was compared to the support vector machine (SVM) model and two state-of-the-art neural networks. An ablation study was performed.The MFFE U-Net achieved a precision of 0.79 ± 0.08, a recall of 0.85 ± 0.11, and an F1-score of 0.82 ± 0.09. Statistically significant improvement was observed when comparing MFFE U-Net with proximity estimation couple SVM (SVM), mU-Net, and Deeplabv3. The HD95 was 2.75 ± 0.44 mm and 3.91 ± 0.83 mm for the 10 patients used in the model construction and 7 patients used for external testing, respectively. The ablation test showed that all five MR sequences contributed to the performance of the final model, with T1ce contributing the most. Convergence analysis, time efficiency analysis, and visualization of the intermediate results further discovered the characteristics of the proposed method.. We present an advanced MFFE learning framework, MFFE U-Net, for effective voxel-wise GBM recurrence prediction. MFFE U-Net performs significantly better than the state-of-the-art networks and can potentially guide early RT intervention of the disease recurrence.
我们旨在开发一种结合干细胞生态位邻近度估计的多模态融合和特征增强 U-Net（MFFE U-Net），以提高脑胶质瘤（GBM）复发的体素预测。从 4 个数据库中回顾性征集了 57 名术前和术后磁共振（MR）扫描患者。术后 MR 扫描包括临床诊断复发前两个月和放射学确认复发当天。通过 T1ce 手动标注复发。首先确定高危复发区域。然后，开发了稀疏多模态特征融合 U-Net。来自 3 个数据库的 50 名患者被分为 70%的训练、10%的验证和 20%的测试。第 4 个机构的 7 名患者用于具有转移学习的外部测试。通过召回率、准确率、F1 评分和 95%分位数的 Hausdorff 距离（HD95）评估模型性能。将所提出的 MFFE U-Net 与支持向量机（SVM）模型和两个最先进的神经网络进行比较。进行了消融研究。MFFE U-Net 的准确率为 0.79 ± 0.08，召回率为 0.85 ± 0.11，F1 评分为 0.82 ± 0.09。与邻近度估计耦合 SVM（SVM）、mU-Net 和 Deeplabv3 相比，MFFE U-Net 具有显著提高。用于模型构建的 10 名患者和用于外部测试的 7 名患者的 HD95 分别为 2.75 ± 0.44mm 和 3.91 ± 0.83mm。消融测试表明，所有五种 MR 序列都有助于最终模型的性能，其中 T1ce 的贡献最大。收敛分析、时间效率分析和中间结果的可视化进一步发现了所提出方法的特点。我们提出了一种先进的 MFFE 学习框架，MFFE U-Net，用于有效的 GBM 复发体素预测。MFFE U-Net 的表现明显优于最先进的网络，有潜力指导疾病复发的早期 RT 干预。
