Department of Radiation Oncology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands.
Department of Radiation Oncology, University of Groningen, University Medical Center Groningen, Groningen, the Netherlands; Machine Learning Lab, Data Science Center in Health (DASH), Groningen, the Netherlands; Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence, University of Groningen, Groningen, the Netherlands.
Radiother Oncol. 2024 Aug;197:110368. doi: 10.1016/j.radonc.2024.110368. Epub 2024 Jun 2.
To optimize our previously proposed TransRP, a model integrating CNN (convolutional neural network) and ViT (Vision Transformer) designed for recurrence-free survival prediction in oropharyngeal cancer and to extend its application to the prediction of multiple clinical outcomes, including locoregional control (LRC), Distant metastasis-free survival (DMFS) and overall survival (OS).
Data was collected from 400 patients (300 for training and 100 for testing) diagnosed with oropharyngeal squamous cell carcinoma (OPSCC) who underwent (chemo)radiotherapy at University Medical Center Groningen. Each patient's data comprised pre-treatment PET/CT scans, clinical parameters, and clinical outcome endpoints, namely LRC, DMFS and OS. The prediction performance of TransRP was compared with CNNs when inputting image data only. Additionally, three distinct methods (m1-3) of incorporating clinical predictors into TransRP training and one method (m4) that uses TransRP prediction as one parameter in a clinical Cox model were compared.
TransRP achieved higher test C-index values of 0.61, 0.84 and 0.70 than CNNs for LRC, DMFS and OS, respectively. Furthermore, when incorporating TransRP's prediction into a clinical Cox model (m4), a higher C-index of 0.77 for OS was obtained. Compared with a clinical routine risk stratification model of OS, our model, using clinical variables, radiomics and TransRP prediction as predictors, achieved larger separations of survival curves between low, intermediate and high risk groups.
TransRP outperformed CNN models for all endpoints. Combining clinical data and TransRP prediction in a Cox model achieved better OS prediction.
为了优化我们之前提出的 TransRP，一种结合了 CNN（卷积神经网络）和 ViT（Vision Transformer）的模型，旨在预测口咽癌无复发生存率，并将其应用扩展到多个临床结局的预测，包括局部区域控制（LRC）、远处无转移生存率（DMFS）和总生存率（OS）。
数据来自于在格罗宁根大学医学中心接受（放）化疗的 400 名诊断为口咽鳞状细胞癌（OPSCC）的患者（300 名用于训练，100 名用于测试）。每位患者的数据包括治疗前的 PET/CT 扫描、临床参数和临床结局终点，即 LRC、DMFS 和 OS。将 TransRP 的预测性能与仅输入图像数据的 CNN 进行了比较。此外，比较了将临床预测因子纳入 TransRP 训练的三种不同方法（m1-3）和一种使用 TransRP 预测作为临床 Cox 模型中一个参数的方法（m4）。
TransRP 对 LRC、DMFS 和 OS 的测试 C 指数值分别为 0.61、0.84 和 0.70，均高于 CNN。此外，当将 TransRP 的预测纳入临床 Cox 模型（m4）时，OS 的 C 指数达到了 0.77。与 OS 的临床常规风险分层模型相比，我们的模型使用临床变量、放射组学和 TransRP 预测作为预测因子，在低、中、高危组之间的生存曲线分离方面取得了更大的效果。
TransRP 在所有终点上均优于 CNN 模型。在 Cox 模型中结合临床数据和 TransRP 预测可实现更好的 OS 预测。
