PET Brain Data
收藏资源简介:
PET脑部数据集由北京师范大学和北京师范大学教育部射线技术重点实验室创建,包含1224名患者的原始PET脑部数据。数据集使用西门子Biograph Vision PET/CT扫描仪采集,每位患者进行120秒的脑部扫描。数据集用于模拟低剂量PET条件,通过缩短扫描时间(30秒、12秒和5秒)来重建图像,旨在解决低剂量PET图像噪声增加的问题,提高图像质量和诊断准确性。
The PET brain dataset was developed by Beijing Normal University and the Key Laboratory of Radiation Technology, Ministry of Education, Beijing Normal University. It encompasses raw PET brain data from 1224 patients. The data was acquired using a Siemens Biograph Vision PET/CT scanner, with each patient receiving a 120-second brain scan. This dataset is designed to simulate low-dose PET scenarios: images are reconstructed by reducing the scan duration to 30 seconds, 12 seconds, and 5 seconds, with the goal of mitigating the elevated noise issue in low-dose PET images and enhancing both image quality and diagnostic accuracy.

- 1Cycle-Constrained Adversarial Denoising Convolutional Network for PET Image Denoising: Multi-Dimensional Validation on Large Datasets with Reader Study and Real Low-Dose Data北京师范大学 · 2024年



