Indrakanti Ashraya Kumar, Wasserthal Jakob, Segeroth Martin, Yang Shan, Nicoli Andrew Phillip, Schulze-Zachau Victor, Lieb Johanna, Cyriac Joshy, Bach Michael, Psychogios Marios, Mutke Matthias Anthony
Department of Diagnostic and Interventional Neuroradiology, Basel University Hospital, Petersgraben 4, 4031, Basel, Switzerland.
Clinic of Radiology and Nuclear Medicine, University Hospital Basel, Petersgraben 4, 4031, Basel, Switzerland.
J Imaging Inform Med. 2025 May 12. doi: 10.1007/s10278-025-01533-3.
The aim of this study was to develop an open-source nnU-Net-based AI model for combined detection and segmentation of unruptured intracranial aneurysms (UICA) in 3D TOF-MRI and compare models trained on datasets with aneurysm-like differential diagnoses. This retrospective study (2020-2023) included 385 anonymized 3D TOF-MRI images from 345 patients (mean age 59 years, 60% female) at multiple centers plus 113 subjects from the ADAM challenge. Images featured untreated or possible UICA and differential diagnoses. Four distinct training datasets were created, and the nnU-Net framework was used for model development. Performance was assessed on a separate test set using sensitivity and false positive (FP)/case rate for detection and DICE score and NSD (normalized surface distance, 0.5 mm threshold) for segmentation. Segmentation performance on the test set was also compared to a second human reader. The four models achieved overall sensitivity between 82 and 85% and an FP/case rate of 0.20 to 0.31, with no significant differences (p = 0.90 and p = 0.16) between them. The primary model showed 85% sensitivity and 0.23 FP/case rate, outperforming the ADAM-challenge winner (61%) and a nnU-Net trained on ADAM data (51%) in sensitivity (p < 0.05). Mean DICE (0.73) and NSD (0.84 for 0.5 mm threshold) for correctly detected UICA did not significantly differ from human reader performance. Our open-source, nnU-Net-based AI model (available at https://zenodo.org/records/13386859 ) demonstrates high sensitivity, low FP rates, and consistent segmentation accuracy for UICA detection and segmentation in 3D TOF-MRI, suggesting its potential to improve clinical diagnosis and monitoring of UICA.
本研究的目的是开发一种基于nnU-Net的开源人工智能模型，用于在三维时间飞跃磁共振血管造影（3D TOF-MRI）中联合检测和分割未破裂颅内动脉瘤（UICA），并比较在具有动脉瘤样鉴别诊断的数据集上训练的模型。这项回顾性研究（2020 - 2023年）纳入了来自多个中心的345例患者（平均年龄59岁，60%为女性）的385份匿名3D TOF-MRI图像，以及来自ADAM挑战赛的113名受试者。图像的特征为未经治疗的或可能的UICA以及鉴别诊断。创建了四个不同的训练数据集，并使用nnU-Net框架进行模型开发。在一个单独的测试集上，使用检测的灵敏度和假阳性（FP）/病例率以及分割的DICE评分和NSD（归一化表面距离，阈值0.5毫米）来评估性能。测试集上的分割性能也与第二位人类阅片者进行了比较。这四个模型的总体灵敏度在82%至85%之间，FP/病例率为0.20至0.31，它们之间无显著差异（p = 0.90和p = 0.16）。主要模型的灵敏度为85%，FP/病例率为0.23，在灵敏度方面优于ADAM挑战赛获胜者（61%）和在ADAM数据上训练的nnU-Net（51%）（p < 0.05）。正确检测到的UICA的平均DICE（0.73）和NSD（0.5毫米阈值时为0.84）与人类阅片者的表现无显著差异。我们基于nnU-Net的开源人工智能模型（可在https://zenodo.org/records/13386859获取）在3D TOF-MRI中对UICA的检测和分割显示出高灵敏度、低FP率和一致的分割准确性，表明其在改善UICA临床诊断和监测方面的潜力。
