Chandrashekar Anirudh, Handa Ashok, Ward Joel, Grau Vicente, Lee Regent
Nuffield Department of Surgical Sciences, John Radcliffe Hospital, University of Oxford, Level 6, Headley Way, Headington, Oxford, OX3 9DU, UK.
Department of Engineering Science, University of Oxford, Oxford, UK.
Insights Imaging. 2022 Mar 14;13(1):45. doi: 10.1186/s13244-022-01161-3.
Positron emission tomography (PET) imaging is a costly tracer-based imaging modality used to visualise abnormal metabolic activity for the management of malignancies. The objective of this study is to demonstrate that non-contrast CTs alone can be used to differentiate regions with different Fluorodeoxyglucose (FDG) uptake and simulate PET images to guide clinical management.
Paired FDG-PET and CT images (n = 298 patients) with diagnosed head and neck squamous cell carcinoma (HNSCC) were obtained from The cancer imaging archive. Random forest (RF) classification of CT-derived radiomic features was used to differentiate metabolically active (tumour) and inactive tissues (ex. thyroid tissue). Subsequently, a deep learning generative adversarial network (GAN) was trained for this CT to PET transformation task without tracer injection. The simulated PET images were evaluated for technical accuracy (PERCIST v.1 criteria) and their ability to predict clinical outcome [(1) locoregional recurrence, (2) distant metastasis and (3) patient survival].
From 298 patients, 683 hot spots of elevated FDG uptake (elevated SUV, 6.03 ± 1.71) were identified. RF models of intensity-based CT-derived radiomic features were able to differentiate regions of negligible, low and elevated FDG uptake within and surrounding the tumour. Using the GAN-simulated PET image alone, we were able to predict clinical outcome to the same accuracy as that achieved using FDG-PET images.
This pipeline demonstrates a deep learning methodology to simulate PET images from CT images in HNSCC without the use of radioactive tracer. The same pipeline can be applied to other pathologies that require PET imaging.
正电子发射断层扫描（PET）成像是一种基于示踪剂的昂贵成像方式，用于可视化异常代谢活动以管理恶性肿瘤。本研究的目的是证明仅使用非增强CT即可区分具有不同氟脱氧葡萄糖（FDG）摄取的区域，并模拟PET图像以指导临床管理。
从癌症影像存档中获取了298例诊断为头颈部鳞状细胞癌（HNSCC）的成对FDG-PET和CT图像。利用基于CT的放射组学特征的随机森林（RF）分类来区分代谢活跃（肿瘤）和不活跃组织（如甲状腺组织）。随后，针对此无需注射示踪剂的CT到PET转换任务训练了一个深度学习生成对抗网络（GAN）。对模拟的PET图像进行技术准确性（PERCIST v.1标准）及其预测临床结果的能力评估（（1）局部区域复发，（2）远处转移和（3）患者生存）。
在298例患者中，识别出683个FDG摄取升高的热点（SUV升高，6.03±1.71）。基于强度的CT衍生放射组学特征的RF模型能够区分肿瘤内部和周围FDG摄取可忽略不计、低和升高的区域。仅使用GAN模拟的PET图像，我们就能以与使用FDG-PET图像相同的准确性预测临床结果。
该流程展示了一种深度学习方法，可在不使用放射性示踪剂的情况下从HNSCC的CT图像模拟PET图像。相同的流程可应用于其他需要PET成像的病理情况。
