Department of Neurosurgery, Xuanwu Hospital, Capital Medical University, Beijing, China; China International Neuroscience Institute, Beijing, China.
Department of Radiology and Nuclear Medicine, Xuanwu Hospital, Capital Medical University, Beijing, China; Beijing Key Laboratory of Magnetic Resonance Imaging and Brain Informatics, Beijing, China.
Eur J Radiol. 2024 Sep;178:111572. doi: 10.1016/j.ejrad.2024.111572. Epub 2024 Jun 13.
Accurate nidus segmentation and quantification have long been challenging but important tasks in the clinical management of Cerebral Arteriovenous Malformation (CAVM). However, there are still dilemmas in nidus segmentation, such as difficulty defining the demarcation of the nidus, observer-dependent variation and time consumption. The aim of this study isto develop an artificial intelligence model to automatically segment the nidus on Time-Of-Flight Magnetic Resonance Angiography (TOF-MRA) images.
A total of 92patients with CAVM who underwent both TOF-MRA and DSA examinations were enrolled. Two neurosurgeonsmanually segmented the nidusonTOF-MRA images,which were regarded as theground-truth reference. AU-Net-basedAImodelwascreatedfor automatic nidus detectionand segmentationonTOF-MRA images.
The meannidus volumes of the AI segmentationmodeland the ground truthwere 5.427 ± 4.996 and 4.824 ± 4.567 mL,respectively. The meandifference in the nidus volume between the two groups was0.603 ± 1.514 mL,which wasnot statisticallysignificant (P = 0.693). The DSC,precision and recallofthe testset were 0.754 ± 0.074, 0.713 ± 0.102 and 0.816 ± 0.098, respectively. The linear correlation coefficient of the nidus volume betweenthesetwo groupswas 0.988, p < 0.001.
The performance of the AI segmentationmodel is moderate consistent with that of manual segmentation. This AI model has great potential in clinical settings, such as preoperative planning, treatment efficacy evaluation, riskstratification and follow-up.
准确的病灶分割和定量一直是脑动静脉畸形（CAVM）临床管理中的一项具有挑战性但很重要的任务。然而，在病灶分割中仍然存在一些难题，例如病灶边界难以界定、观察者间存在差异和耗费时间等。本研究旨在开发一种人工智能模型，以便自动对时间飞越磁共振血管造影（TOF-MRA）图像中的病灶进行分割。
共纳入 92 例接受 TOF-MRA 和 DSA 检查的 CAVM 患者。由两名神经外科医生手动对 TOF-MRA 图像中的病灶进行分割，将其作为ground-truth 参考。基于 AU-Net 的 AI 模型用于自动检测和分割 TOF-MRA 图像中的病灶。
AI 分割模型和 ground-truth 的平均病灶体积分别为 5.427±4.996 和 4.824±4.567 mL。两组间病灶体积的平均差异为 0.603±1.514 mL，差异无统计学意义（P=0.693）。测试集的 DSC、精度和召回率分别为 0.754±0.074、0.713±0.102 和 0.816±0.098。两组间病灶体积的线性相关系数为 0.988，p<0.001。
AI 分割模型的性能与手动分割具有中等一致性。该 AI 模型在术前规划、治疗效果评估、风险分层和随访等临床环境中具有很大的应用潜力。
