Park Jiyeon, Lim Chae Young, Won So Yeon, Na Han Kyu, Lee Phil Hyu, Baek Sun-Young, Roh Yun Hwa, Seong Minjung, Sim Yongsik, Kim Eung Yeop, Kim Sung Tae, Sohn Beomseok
Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea.
Department of Neurology, Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea.
Korean J Radiol. 2025 Aug;26(8):771-781. doi: 10.3348/kjr.2025.0208.
To evaluate the effect of deep learning (DL)-based artificial intelligence (AI) software on the diagnostic performance of radiologists with different experience levels in detecting nigrosome 1 (N1) abnormalities on susceptibility map-weighted imaging (SMwI).
This retrospective diagnostic case-control study analyzed 139 SMwI scans of 59 patients with Parkinson's disease (PD) and 80 healthy participants. Participants were imaged using 3T MRI, and AI-generated assessments for N1 abnormalities were obtained using an AI model (version 1.0.1.0; Heuron Corporation, Seoul, Korea), which utilized YOLOX-based object detection and SparseInst segmentation models. Four radiologists (two experienced neuroradiologists and two less experienced residents) evaluated N1 abnormalities with and without AI in a crossover study design. Diagnostic performance metrics, inter-reader agreements, and reader responses to AI-generated assessments were evaluated.
Use of AI significantly improved diagnostic performance compared with interpretation without it across three readers, with significant increases in specificity (0.86 vs. 0.94, = 0.004; 0.91 vs. 0.97, = 0.024; and 0.90 vs. 0.97, = 0.012). Inter-reader agreement also improved with AI, as Fleiss's kappa increased from 0.73 (95% confidence interval [CI]: 0.61-0.84) to 0.87 (95% CI: 0.76-0.99). The net reclassification index (NRI) demonstrated significant improvement in three of the four readers. When grouped by experience level, less experienced readers showed greater improvement (NRI = 12.8%, 95% CI: 0.067-0.190) than experienced readers (NRI = 0.8%, 95% CI: -0.037-0.051). In the less experienced group, reader-AI disagreement was significantly higher in the PD group than in the normal group (8.1% vs. 3.8%, = 0.029).
DL-based AI enhances the diagnostic performance in detecting N1 abnormalities on SMwI, particularly benefiting less experienced radiologists. These findings underscore the potential for improving diagnostic workflows for PD.
评估基于深度学习（DL）的人工智能（AI）软件对不同经验水平的放射科医生在检测黑质小体1（N1）在磁化率图加权成像（SMwI）上异常情况时诊断性能的影响。
这项回顾性诊断病例对照研究分析了59例帕金森病（PD）患者和80名健康参与者的139次SMwI扫描。参与者使用3T磁共振成像（MRI）进行成像，并使用一种AI模型（版本1.0.1.0；韩国首尔的Heuron公司）获得关于N1异常的AI生成评估，该模型利用基于YOLOX的目标检测和SparseInst分割模型。在交叉研究设计中，四名放射科医生（两名经验丰富的神经放射科医生和两名经验较少的住院医生）在有和没有AI辅助的情况下评估N1异常情况。评估了诊断性能指标、阅片者间的一致性以及阅片者对AI生成评估的反应。
与没有AI辅助的解读相比，三名阅片者使用AI后诊断性能显著提高，特异性显著增加（分别为0.86对0.94，P = 0.004；0.91对0.97，P = 0.024；0.90对0.97，P = 0.012）。AI辅助下阅片者间的一致性也有所提高，Fleiss卡方值从0.73（95%置信区间[CI]：0.61 - 0.84）增加到0.87（95%CI：0.76 - 0.99）。四名阅片者中有三名的净重新分类指数（NRI）显示出显著改善。按经验水平分组时，经验较少的阅片者比经验丰富的阅片者改善更大（NRI = 12.8%，95%CI：0.067 - 0.190）（经验丰富的阅片者NRI = 0.8%，95%CI： - 0.037 - 0.051）。在经验较少的组中，PD组阅片者与AI的分歧显著高于正常组（8.1%对3.8%，P = 0.029）。
基于DL的AI提高了在SMwI上检测N1异常的诊断性能，尤其使经验较少的放射科医生受益。这些发现强调了改善PD诊断工作流程的潜力。
