Cao Weiguo, Howe Benjamin M, Ramanathan Sumana, Rhodes Nicholas G, Korfiatis Panagiotis, Amrami Kimberly K, Spinner Robert J, Kline Timothy L
Department of Radiology, Mayo Clinic, 200 First Street SW, Charlton 1, Rochester, MN 55905, USA.
Department of Radiology, Mayo Clinic, 200 First Street SW, Charlton 1, Rochester, MN 55905, USA.
Eur J Radiol. 2024 Dec;181:111744. doi: 10.1016/j.ejrad.2024.111744. Epub 2024 Sep 18.
This study aims to seek an optimized deep learning model for differentiating non-traumatic brachial plexopathy from routine MRI scans.
This retrospective study collected patients through the electronic medical records (EMR) or pathological reports at Mayo Clinic and underwent BP MRI from January 2002 to December 2022. Using sagittal T1, fluid-sensitive and post-gadolinium images, a radiology panel selected BP's region of interest (ROI) to form 3 dimensional volumes for this study. We designed six deep learning schemes to conduct BP abnormality differentiation across three MRI sequences. Utilizing five prestigious deep learning networks as the backbone, we trained and validated these models by nested five-fold cross-validation schemes. Furthermore, we defined a 'method score' derived from the radar charts as a quantitative indicator as the guidance of the preference of the best model.
This study selected 196 patients from initial 267 candidates. A total of 256 BP MRI series were compiled from them, comprising 123 normal and 133 abnormal series. The abnormal series included 4 sub-categories, et al. breast cancer (22.5 %), lymphoma (27.1 %), inflammatory conditions (33.1 %) and others (17.2 %). The best-performing model was produced by feature merging mode with triple MRI joint strategy (AUC, 92.2 %; accuracy, 89.5 %) exceeding the multiple channel merging mode (AUC, 89.6 %; accuracy, 89.0 %), solo channel volume mode (AUC, 89.2 %; accuracy, 86.7 %) and the remaining. Evaluated by method score (maximum 2.37), the feature merging mode with backbone of VGG16 yielded the highest score of 1.75 under the triple MRI joint strategy.
Deployment of deep learning models across sagittal T1, fluid-sensitive and post-gadolinium MRI sequences demonstrated great potential for brachial plexopathy diagnosis. Our findings indicate that utilizing feature merging mode and multiple MRI joint strategy may offer satisfied deep learning model for BP abnormalities than solo-sequence analysis.
本研究旨在寻找一种优化的深度学习模型，用于从常规MRI扫描中鉴别非创伤性臂丛神经病变。
本回顾性研究通过梅奥诊所的电子病历（EMR）或病理报告收集患者，这些患者在2002年1月至2022年12月期间接受了臂丛神经MRI检查。利用矢状位T1、液体敏感序列和钆剂增强后图像，一个放射学专家组选择臂丛神经的感兴趣区域（ROI）以形成用于本研究的三维容积。我们设计了六种深度学习方案，以在三个MRI序列上进行臂丛神经异常鉴别。利用五个著名的深度学习网络作为主干，我们通过嵌套五折交叉验证方案对这些模型进行训练和验证。此外，我们将从雷达图得出的“方法分数”定义为定量指标，作为最佳模型偏好的指导。
本研究从最初的267名候选者中选择了196名患者。从中总共汇编了256个臂丛神经MRI系列，包括123个正常系列和133个异常系列。异常系列包括4个亚类，即乳腺癌（22.5%）、淋巴瘤（27.1%）、炎症性疾病（33.1%）和其他（17.2%）。表现最佳的模型是由具有三重MRI联合策略的特征合并模式产生的（AUC，92.2%；准确率，89.5%），超过了多通道合并模式（AUC，89.6%；准确率，89.0%）、单通道容积模式（AUC，89.2%；准确率，86.7%）和其他模式。通过方法分数（最大值2.37）评估，在三重MRI联合策略下，以VGG16为主干的特征合并模式获得了最高分数1.75。
在矢状位T1、液体敏感序列和钆剂增强后MRI序列上部署深度学习模型显示出在臂丛神经病变诊断方面的巨大潜力。我们的研究结果表明，与单序列分析相比，利用特征合并模式和多MRI联合策略可能为臂丛神经异常提供令人满意的深度学习模型。
