College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, Zhejiang, China.
Department of Neurology and Epilepsy Center, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China.
Seizure. 2024 Apr;117:126-132. doi: 10.1016/j.seizure.2024.02.009. Epub 2024 Feb 15.
Focal cortical dysplasia (FCD) is a common etiology of drug-resistant focal epilepsy. Visual identification of FCD is usually time-consuming and depends on personal experience. Herein, we propose an automated type II FCD detection approach utilizing multi-modal data and 3D convolutional neural network (CNN).
MRI and positron emission tomography (PET) data of 82 patients with FCD were collected, including 55 (67.1%) histopathologically, and 27 (32.9%) radiologically diagnosed patients. Three types of morphometric feature maps and three types of tissue maps were extracted from the T1-weighted images. These maps, T1, and PET images formed the inputs for CNN. Five-fold cross-validations were carried out on the training set containing 62 patients, and the model behaving best was chosen to detect FCD on the test set of 20 patients. Furthermore, ablation experiments were performed to estimate the value of PET data and CNN.
On the validation set, FCD was detected in 90.3% of the cases, with an average of 1.7 possible lesions per patient. The sensitivity on the test set was 90.0%, with 1.85 possible lesions per patient. Without the PET data, the sensitivity decreased to 80.0%, and the average lesion number increased to 2.05 on the test set. If an artificial neural network replaced the CNN, the sensitivity decreased to 85.0%, and the average lesion number increased to 4.65.
Automated detection of FCD with high sensitivity and few false-positive findings is feasible based on multi-modal data. PET data and CNN could improve the performance of automated detection.
局灶性皮质发育不良（FCD）是耐药性局灶性癫痫的常见病因。FCD 的视觉识别通常很耗时，并且依赖于个人经验。在此，我们提出了一种利用多模态数据和 3D 卷积神经网络（CNN）的自动 II 型 FCD 检测方法。
收集了 82 例 FCD 患者的 MRI 和正电子发射断层扫描（PET）数据，其中 55 例（67.1%）经组织病理学证实，27 例（32.9%）经影像学诊断。从 T1 加权图像中提取了三种形态特征图和三种组织图。这些图、T1 和 PET 图像构成了 CNN 的输入。在包含 62 例患者的训练集中进行了五折交叉验证，并选择表现最佳的模型来检测 20 例患者的测试集上的 FCD。此外，还进行了消融实验以估计 PET 数据和 CNN 的价值。
在验证集上，90.3%的病例检测到 FCD，平均每位患者有 1.7 个可能的病变。在测试集上的敏感性为 90.0%，平均每位患者有 1.85 个可能的病变。如果没有 PET 数据，敏感性降低至 80.0%，平均病变数量增加到 2.05 个。如果用人工神经网络代替 CNN，敏感性降低至 85.0%，平均病变数量增加到 4.65。
基于多模态数据，实现了具有高灵敏度和较少假阳性发现的 FCD 自动检测是可行的。PET 数据和 CNN 可以提高自动检测的性能。
