Toner Brian, Arberet Simon, Zhang Shu, Han Fei, Ahanonu Eze, Goerke Ute, Johnson Kevin, Abouelfetouh Zeyad, Codreanu Ion, Sridhar Sajeev, Arif-Tiwari Hina, Deshpande Vibhas, Martin Diego R, Nadar Mariappan, Altbach Maria I, Bilgin Ali
Program in Applied Mathematics, The University of Arizona, Tucson, Arizona, USA.
Digital Technology & Innovation, Siemens Healthineers, Princeton, New Jersey, USA.
Magn Reson Med. 2025 Aug 5. doi: 10.1002/mrm.70017.
To accelerate respiratory triggered free-breathing T2 mapping of the abdomen while maintaining high-quality anatomical images, accurate T2 maps, and fast reconstruction times.
We developed a flexible deep learning framework that can be trained in a fully supervised manner to improve T2-weighted images or in a self-supervised manner to reconstruct T2 maps.
For retrospectively undersampled data, anatomical images and T2 maps reconstructed by the proposed deep learning method demonstrated reduced voxel-wise error compared to existing traditional and compressed sensing techniques. Reconstruction times were approximately 1 s per slice, significantly faster than existing compressed sensing techniques. Prospectively undersampled data were also acquired to assess the model.
The proposed deep-learning framework reconstructed high-quality anatomical images and accurate T2 maps from datasets undersampled to only 160 total radial views (5 views per echo time), enabling full liver coverage in under three minutes on average with per-slice reconstruction times of approximately one second.
在保持高质量解剖图像、准确的T2图谱以及快速重建时间的同时，加速腹部呼吸触发自由呼吸T2图谱成像。
我们开发了一个灵活的深度学习框架，该框架可以通过全监督方式进行训练以改善T2加权图像，或者通过自监督方式进行训练以重建T2图谱。
对于回顾性欠采样数据，与现有的传统和压缩感知技术相比，通过所提出的深度学习方法重建的解剖图像和T2图谱显示出体素级误差降低。重建时间约为每切片1秒，明显快于现有的压缩感知技术。还采集了前瞻性欠采样数据以评估该模型。
所提出的深度学习框架能够从仅欠采样到总共160个径向视图（每个回波时间5个视图）的数据集中重建高质量的解剖图像和准确的T2图谱，平均不到三分钟即可实现全肝覆盖，每切片重建时间约为一秒。
