文献速递泌尿生殖腹部MR

基于卷积神经网络的重建加速前列腺T2加权MR成像:一项回顾性与前瞻性研究

Convolutional neural network-based reconstruction for acceleration of prostate T2 weighted MR imaging: a retro- and prospective study.
PubMed 文献 · Br J Radiol · 2022-02-11

研究开发了基于深度神经网络(DNN)的并行成像重建方法,用于前列腺MRI高度加速的二维涡轮自旋回波(TSE)数据。155例参与者数据用于训练和测试,构建了针对采集次数NAQ=1的DNN-N1和NAQ=2的DNN-N2两个模型。结果显示,NAQ=1时扫描时间减少71%,NAQ=2时减少42%;DNN图像定量误差指标优于TSE图像(NRMSE降低29-43%,SSIM提高4-13%,PSNR提高2.8-4.8 dB;p<0.001);DNN-N2在视觉适宜性所有参数上优于或相当于常规TSE(TSE-Conv);病灶检测与TSE-Conv几乎完美一致(κ>0.9)。结论认为DNN重建在高度加速前列腺TSE成像中质量与常规TSE相当,框架将扫描时间减少42%且无需修改序列,具有临床应用潜力。

要点

  • DNN重建可在NAQ=1时减少71%、NAQ=2时减少42%的扫描时间。
  • DNN图像定量误差指标优于TSE图像:NRMSE降低29-43%,SSIM提高4-13%,PSNR提高2.8-4.8 dB(p<0.001)。
  • DNN-N2在视觉适宜性所有参数上优于或相当于常规TSE,病灶检测与常规TSE几乎完美一致(κ>0.9)。

核心结论

  • 基于DNN的重建在高度加速前列腺TSE成像中显示出与常规TSE相当的质量;该框架将常规前列腺TSE成像的扫描时间减少42%且无需修改序列,显示出巨大的临床应用潜力。

研究设计与样本

  • 回顾性与前瞻性研究,共获取155例参与者数据用于训练和测试;根据输入图像的采集次数(NAQ)生成两个DNN模型:DNN-N1用于NAQ=1,DNN-N2用于NAQ=2;在测试数据中比较DNN与TSE图像的定量误差指标,并由两名放射科医生对加速扫描(DNN-N1和DNN-N2)与常规扫描(TSE-Conv)的视觉适宜性进行九项参数评估,通过前列腺影像报告和数据系统评估病灶检测。

诊断标准与关键数值

  • 定量评估显示DNN图像较TSE图像NRMSE降低29-43%、SSIM提高4-13%、PSNR提高2.8-4.8 dB(p<0.001);病灶检测中DNN图像与TSE-Conv几乎完美一致(κ>0.9)。

临床应用与阅片要点

  • 作者指出该框架将常规前列腺TSE成像的扫描时间减少42%且无需修改序列,显示出巨大的临床应用潜力。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

OBJECTIVE: The aim of this study was to develop a deep neural network (DNN)-based parallel imaging reconstruction for highly accelerated 2D turbo spin echo (TSE) data in prostate MRI without quality degradation compared to conventional scans.

METHODS: 155 participant data were acquired for training and testing. Two DNN models were generated according to the number of acquisitions (NAQ) of input images: DNN-N1 for NAQ = 1 and DNN-N2 for NAQ = 2. In the test data, DNN and TSE images were compared by quantitative error metrics. The visual appropriateness of DNN reconstructions on accelerated scans (DNN-N1 and DNN-N2) and conventional scans (TSE-Conv) was assessed for nine parameters by two radiologists. The lesion detection was evaluated at DNNs and TES-Conv by prostate imaging-reporting and data system.

RESULTS: The scan time was reduced by 71% at NAQ = 1, and 42% at NAQ = 2. Quantitative evaluation demonstrated the better error metrics of DNN images (29-43% lower NRMSE, 4-13% higher structure similarity index, and 2.8-4.8 dB higher peak signal-to-noise ratio; p < 0.001) than TSE images. In the assessment of the visual appropriateness, both radiologists evaluated that DNN-N2 showed better or comparable performance in all parameters compared to TSE-Conv. In the lesion detection, DNN images showed almost perfect agreement (κ > 0.9) scores with TSE-Conv.

CONCLUSIONS: DNN-based reconstruction in highly accelerated prostate TSE imaging showed comparable quality to conventional TSE.

ADVANCES IN KNOWLEDGE: Our framework reduces the scan time by 42% of conventional prostate TSE imaging without sequence modification, revealing great potential for clinical application.

来源:Br J Radiol,查看原文。版权归原作者所有。

引用

  1. Jung, Kim, Ko, et al. Convolutional neural network-based reconstruction for acceleration of prostate T2 weighted MR imaging: a retro- and prospective study. Br J Radiol. 2022. PMID: 35148172. doi:10.1259/bjr.20211378

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。