Starke Sebastian, Zwanenburg Alex, Leger Karoline, Lohaus Fabian, Linge Annett, Kalinauskaite Goda, Tinhofer Inge, Guberina Nika, Guberina Maja, Balermpas Panagiotis, Grün Jens von der, Ganswindt Ute, Belka Claus, Peeken Jan C, Combs Stephanie E, Boeke Simon, Zips Daniel, Richter Christian, Troost Esther G C, Krause Mechthild, Baumann Michael, Löck Steffen
Helmholtz-Zentrum Dresden-Rossendorf, Department of Information Services and Computing, 01328 Dresden, Germany.
OncoRay-National Center for Radiation Research in Oncology, Faculty of Medicine and University Hospital Carl Gustav Carus, Technische Universität Dresden, Helmholtz-Zentrum Dresden-Rossendorf, 01309 Dresden, Germany.
Cancers (Basel). 2023 Oct 9;15(19):4897. doi: 10.3390/cancers15194897.
Neural-network-based outcome predictions may enable further treatment personalization of patients with head and neck cancer. The development of neural networks can prove challenging when a limited number of cases is available. Therefore, we investigated whether multitask learning strategies, implemented through the simultaneous optimization of two distinct outcome objectives (multi-outcome) and combined with a tumor segmentation task, can lead to improved performance of convolutional neural networks (CNNs) and vision transformers (ViTs). Model training was conducted on two distinct multicenter datasets for the endpoints loco-regional control (LRC) and progression-free survival (PFS), respectively. The first dataset consisted of pre-treatment computed tomography (CT) imaging for 290 patients and the second dataset contained combined positron emission tomography (PET)/CT data of 224 patients. Discriminative performance was assessed by the concordance index (C-index). Risk stratification was evaluated using log-rank tests. Across both datasets, CNN and ViT model ensembles achieved similar results. Multitask approaches showed favorable performance in most investigations. Multi-outcome CNN models trained with segmentation loss were identified as the optimal strategy across cohorts. On the PET/CT dataset, an ensemble of multi-outcome CNNs trained with segmentation loss achieved the best discrimination (C-index: 0.29, 95% confidence interval (CI): 0.22-0.36) and successfully stratified patients into groups with low and high risk of disease progression (p=0.003). On the CT dataset, ensembles of multi-outcome CNNs and of single-outcome ViTs trained with segmentation loss performed best (C-index: 0.26 and 0.26, CI: 0.18-0.34 and 0.18-0.35, respectively), both with significant risk stratification for LRC in independent validation (p=0.002 and p=0.011). Further validation of the developed multitask-learning models is planned based on a prospective validation study, which has recently completed recruitment.
基于神经网络的预后预测可能会使头颈癌患者的治疗更加个性化。当可用病例数量有限时，神经网络的开发可能会面临挑战。因此，我们研究了通过同时优化两个不同的预后目标（多预后）并结合肿瘤分割任务来实施的多任务学习策略，是否能够提高卷积神经网络（CNN）和视觉Transformer（ViT）的性能。分别在两个不同的多中心数据集上进行模型训练，以局部区域控制（LRC）和无进展生存期（PFS）为终点。第一个数据集由290例患者的治疗前计算机断层扫描（CT）成像组成，第二个数据集包含224例患者的正电子发射断层扫描（PET）/CT联合数据。通过一致性指数（C指数）评估判别性能。使用对数秩检验评估风险分层。在两个数据集中，CNN和ViT模型集成取得了相似的结果。多任务方法在大多数研究中表现出良好的性能。在各个队列中，使用分割损失训练的多预后CNN模型被确定为最佳策略。在PET/CT数据集上，使用分割损失训练的多预后CNN集成实现了最佳判别（C指数：0.29，95%置信区间（CI）：0.22-0.36），并成功将患者分为疾病进展低风险和高风险组（p=0.003）。在CT数据集上，使用分割损失训练的多预后CNN和单预后ViT集成表现最佳（C指数分别为0.26和0.26，CI分别为0.18-0.34和0.18-0.35），两者在独立验证中对LRC均有显著的风险分层（p=0.002和p=0.011）。计划基于一项前瞻性验证研究对所开发的多任务学习模型进行进一步验证，该研究最近已完成招募。
