用于前列腺评估的人工智能生成的表观扩散系数（AI-ADC）图：一项多阅片者研究。
Artificial intelligence-generated apparent diffusion coefficient (AI-ADC) maps for prostate gland assessment: a multi-reader study.
作者信息
Ozyoruk Kutsev Bengisu, Harmon Stephanie A, Yilmaz Enis C, Huang Erich P, Gelikman David G, Gaur Sonia, Giganti Francesco, Law Yan Mee, Margolis Daniel J, Jadda Pavan Kumar, Raavi Sitarama, Gurram Sandeep, Wood Bradford J, Pinto Peter A, Choyke Peter L, Turkbey Baris
机构信息
Artificial Intelligence Resource, Molecular Imaging Branch, National Cancer Institute, Bethesda, MD, USA.
Molecular Imaging Branch, National Cancer Institute, Bethesda, MD, USA.
出版信息
Eur Radiol. 2025 Jul 21. doi: 10.1007/s00330-025-11871-z.
OBJECTIVE
To compare the quality of AI-ADC maps and standard ADC maps in a multi-reader study.
MATERIALS AND METHODS
Multi-reader study included 74 consecutive patients (median age = 66 years, [IQR = 57.25-71.75 years]; median PSA = 4.30 ng/mL [IQR = 1.33-7.75 ng/mL]) with suspected or confirmed PCa, who underwent mpMRI between October 2023 and January 2024. The study was conducted in two rounds, separated by a 4-week wash-out period. In each round, four readers evaluated T2W-MRI and standard or AI-generated ADC (AI-ADC) maps. Fleiss' kappa, quadratic-weighted Cohen's kappa statistics were used to assess inter-reader agreement. Linear mixed effect models were employed to compare the quality evaluation of standard versus AI-ADC maps.
RESULTS
AI-ADC maps exhibited significantly enhanced imaging quality compared to standard ADC maps with higher ratings in windowing ease (β = 0.67 [95% CI 0.30-1.04], p < 0.05), prostate boundary delineation (β = 1.38 [95% CI 1.03-1.73], p < 0.001), reductions in distortion (β = 1.68 [95% CI 1.30-2.05], p < 0.001), noise (β = 0.56 [95% CI 0.24-0.88], p < 0.001). AI-ADC maps reduced reacquisition requirements for all readers (β = 2.23 [95% CI 1.69-2.76], p < 0.001), supporting potential workflow efficiency gains. No differences were observed between AI-ADC and standard ADC maps' inter-reader agreement.
CONCLUSION
Our multi-reader study demonstrated that AI-ADC maps improved prostate boundary delineation, had lower image noise, fewer distortions, and higher overall image quality compared to ADC maps.
KEY POINTS
Question Can we synthesize apparent diffusion coefficient (ADC) maps with AI to achieve higher quality maps? Findings On average, readers rated quality factors of AI-ADC maps higher than ADC maps in 34.80% of cases, compared to 5.07% for ADC (p < 0.01). Clinical relevance AI-ADC maps may serve as a reliable diagnostic support tool thanks to their high quality, particularly when the acquired ADC maps include artifacts.
目的
在一项多阅片者研究中比较人工智能生成的表观扩散系数（AI-ADC）图和标准ADC图的质量。
材料与方法
多阅片者研究纳入了2023年10月至2024年1月期间连续74例疑似或确诊前列腺癌的患者（中位年龄 = 66岁，[四分位间距（IQR）= 57.25 - 71.75岁]；中位前列腺特异性抗原（PSA）= 4.30 ng/mL [IQR = 1.33 - 7.75 ng/mL]），这些患者均接受了多参数磁共振成像（mpMRI）检查。该研究分两轮进行，两轮之间有4周的洗脱期。在每一轮中，四名阅片者对T2加权磁共振成像（T2W-MRI）以及标准或人工智能生成的ADC（AI-ADC）图进行评估。采用弗莱iss卡方检验、二次加权科恩卡方统计量来评估阅片者间的一致性。使用线性混合效应模型比较标准ADC图与AI-ADC图的质量评估。
结果
与标准ADC图相比，AI-ADC图的成像质量显著提高，在窗宽调整的便利性方面评分更高（β = 0.67 [95%置信区间（CI）0.30 - 1.04]，p < 0.05），前列腺边界勾画方面（β = 1.38 [95% CI 1.03 - 1.73]，p < 0.001），失真减少方面（β = 1.68 [95% CI 1.30 - 2.05]，p < 0.001），噪声方面（β = 0.56 [95% CI 0.24 - 0.88]，p < 0.001）。AI-ADC图减少了所有阅片者的重新采集需求（β = 2.23 [95% CI 1.69 - 2.76]，p < 0.001），这表明其可能提高工作流程效率。在AI-ADC图和标准ADC图的阅片者间一致性方面未观察到差异。
结论
我们的多阅片者研究表明，与ADC图相比，AI-ADC图改善了前列腺边界勾画，图像噪声更低，失真更少，整体图像质量更高。
关键点
问题我们能否利用人工智能合成表观扩散系数（ADC）图以获得更高质量的图？研究结果平均而言，在34.80%的病例中，阅片者对AI-ADC图质量因素的评分高于ADC图，而对ADC图质量因素评分高于AI-ADC图的病例占5.07%（p < 0.01）。临床意义由于AI-ADC图质量高，特别是当采集的ADC图包含伪影时，其可作为一种可靠的诊断支持工具。
Zotero 插件