Developing Brain Institute, Division of Diagnostic Imaging and Radiology, Children's National Hospital, Washington, District of Columbia, USA.
Department of Neurology, Children's National Hospital, Washington, District of Columbia, USA.
J Magn Reson Imaging. 2021 Sep;54(3):818-829. doi: 10.1002/jmri.27649. Epub 2021 Apr 23.
Due to random motion of fetuses and maternal respirations, image quality of fetal brain MRIs varies considerably. To address this issue, visual inspection of the images is performed during acquisition phase and after 3D-reconstruction, and the images are re-acquired if they are deemed to be of insufficient quality. However, this process is time-consuming and subjective. Multi-instance (MI) deep learning methods (DLMs) may perform this task automatically.
To propose an MI count-based DLM (MI-CB-DLM), an MI vote-based DLM (MI-VB-DLM), and an MI feature-embedding DLM (MI-FE-DLM) for automatic assessment of 3D fetal-brain MR image quality. To quantify influence of fetal gestational age (GA) on DLM performance.
Retrospective.
Two hundred and seventy-one MR exams from 211 fetuses (mean GA ± SD = 30.9 ± 5.5 weeks).
FIELD STRENGTH/SEQUENCE: T -weighted single-shot fast spin-echo acquired at 1.5 T.
The T -weighted images were reconstructed in 3D. Then, two fetal neuroradiologists, a clinical neuroscientist, and a fetal MRI technician independently labeled the reconstructed images as 1 or 0 based on image quality (1 = high; 0 = low). These labels were fused and served as ground truth. The proposed DLMs were trained and evaluated using three repeated 10-fold cross-validations (training and validation sets of 244 and 27 scans). To quantify GA influence, this variable was included as an input of the DLMs.
DLM performance was evaluated using precision, recall, F-score, accuracy, and AUC values.
Precision, recall, F-score, accuracy, and AUC averaged over the three cross validations were 0.85 ± 0.01, 0.85 ± 0.01, 0.85 ± 0.01, 0.85 ± 0.01, 0.93 ± 0.01, for MI-CB-DLM (without GA); 0.75 ± 0.03, 0.75 ± 0.03, 0.75 ± 0.03, 0.75 ± 0.03, 0.81 ± 0.03, for MI-VB-DLM (without GA); 0.81 ± 0.01, 0.81 ± 0.01, 0.81 ± 0.01, 0.81 ± 0.01, 0.89 ± 0.01, for MI-FE-DLM (without GA); and 0.86 ± 0.01, 0.86 ± 0.01, 0.86 ± 0.01, 0.86 ± 0.01, 0.93 ± 0.01, for MI-CB-DLM with GA.
MI-CB-DLM performed better than other DLMs. Including GA as an input of MI-CB-DLM improved its performance. MI-CB-DLM may potentially be used to objectively and rapidly assess fetal MR image quality.
4 TECHNICAL EFFICACY: Stage 3.
由于胎儿和母体呼吸的随机运动，胎儿脑部 MRI 的图像质量差异很大。为了解决这个问题，在采集阶段和三维重建后对图像进行视觉检查，如果认为图像质量不足，则重新采集图像。然而，这个过程既耗时又主观。多实例（MI）深度学习方法（DLM）可以自动执行此任务。
提出一种基于 MI 计数的 DLM（MI-CB-DLM）、一种基于 MI 投票的 DLM（MI-VB-DLM）和一种基于 MI 特征嵌入的 DLM（MI-FE-DLM），用于自动评估 3D 胎儿脑部 MRI 图像质量。定量研究胎儿胎龄（GA）对 DLM 性能的影响。
回顾性研究。
211 名胎儿的 271 项 MRI 检查（平均 GA ± SD=30.9 ± 5.5 周）。
磁场强度/序列：1.5T 下的 T2 加权单次激发快速自旋回波。
重建的 T2 加权图像在 3D 中重建。然后，两名胎儿神经放射科医生、一名临床神经科学家和一名胎儿 MRI 技师根据图像质量独立地将重建的图像标记为 1 或 0（1=高；0=低）。这些标签被融合并作为地面实况。使用三次重复的 10 折交叉验证（训练集和验证集分别为 244 次和 27 次扫描）来训练和评估所提出的 DLM。为了量化 GA 的影响，将该变量作为 DLM 的输入。
使用精度、召回率、F 分数、准确性和 AUC 值评估 DLM 性能。
三次交叉验证的平均精度、召回率、F 分数、准确性和 AUC 分别为 0.85±0.01、0.85±0.01、0.85±0.01、0.85±0.01、0.93±0.01，用于 MI-CB-DLM（无 GA）；0.75±0.03、0.75±0.03、0.75±0.03、0.75±0.03、0.81±0.03，用于 MI-VB-DLM（无 GA）；0.81±0.01、0.81±0.01、0.81±0.01、0.81±0.01、0.89±0.01，用于 MI-FE-DLM（无 GA）；和 0.86±0.01、0.86±0.01、0.86±0.01、0.86±0.01、0.93±0.01，用于 MI-CB-DLM 与 GA。
MI-CB-DLM 的性能优于其他 DLM。将 GA 作为 MI-CB-DLM 的输入可以提高其性能。MI-CB-DLM 可能具有潜力，用于客观、快速地评估胎儿 MRI 图像质量。
4 级 技术功效：3 级。
