合成医学影像的评估指标系统综述
本系统综述依据PRISMA指南,检索PubMed/MEDLINE、EMBASE、Scopus和arXiv中2015年至2025年4月30日发表的研究,最终纳入47项研究,系统总结了用于评价合成医学影像保真度、真实性、多样性和临床有效性的指标。结果显示评估实践高度异质,最常用指标为PSNR、SSIM、MAE和FID,提示缺乏单一指标能同时覆盖所有维度,需建立任务导向的分层评估框架。
要点
- 合成影像评估应结合保真度、真实性、多样性和临床有效性多个维度,不可依赖单一指标。
- PSNR、SSIM、MAE和FID是最常用指标,但指标选择常受数据可用性影响而非临床目的驱动。
- 报告合成影像研究结果时,应说明所选评估指标的适用维度和局限性,提高研究间可比性。
核心结论
- 合成医学影像评估策略存在显著异质性,没有任何单一指标能同时捕获保真度、真实性、多样性和临床有效性。
- 指标选择常由数据可用性决定,而非由临床目的驱动。
- 建议建立任务导向的分层评估框架,以提高研究可比性并促进临床转化应用。
研究设计与样本
- 系统综述,依据PRISMA指南开展。
- 检索PubMed/MEDLINE、EMBASE、Scopus和arXiv,时间范围为2015年至2025年4月30日,并补充纳入文献的引文筛选。
- 纳入标准为应用或提出指标来评估合成医学影像保真度、真实性、多样性和临床有效性的全文研究。
诊断标准与关键数值
- 纳入47项研究,专家评估(n=25, 53%)与基于参考的评估最常见(n=25, 53%)。
- 无参考指标评估(n=24, 51%)与基于任务的评估(n=24, 51%)的使用频率相当。
- 最常用个体指标为PSNR(n=16, 34%)、SSIM(n=15, 32%)、MAE(n=12, 26%)和FID(n=12, 26%)。
临床应用与阅片要点
- 评估合成医学影像时,应根据具体临床任务选择相应维度的多个指标,而非仅用PSNR或SSIM等像素级指标。
- 当数据条件受限时,应明确说明所选指标的适用维度范围,避免将单一指标结果外推至临床有效性。
- 采用分层评估策略,将保真度、真实性、多样性和临床有效性分开评价,有助于明确每类指标的实际意义。
局限性
- 原文未提供纳入研究的方法学质量评估细节,可能影响结论的证据强度。
- 纳入研究之间的评估条件与数据特征差异较大,限制了指标间的直接比较。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
BACKGROUND: Advances in generative artificial intelligence (AI) have accelerated the development and application of synthetic medical imaging. Despite this rapid progress, the evaluation of synthetic medical images remains heterogeneous, with numerous metrics proposed to assess fidelity, realism, diversity, and clinical validity. Currently, no standardized framework exists to guide the selection, interpretation, or comparison of these metrics, limiting reproducibility and cross-study comparability. This systematic review aims to comprehensively summarize and categorize existing metrics used to assess these complementary dimensions of synthetic medical images.
METHODS: A systematic review was conducted in accordance with PRISMA guidelines. PubMed/MEDLINE, EMBASE, Scopus, and arXiv were searched for studies published between 2015 and April 30, 2025, supplemented by citation screening of included studies. Eligible studies were full-text articles that applied or proposed metrics to evaluate the fidelity, realism, diversity, and/or clinical validity in synthetic medical images.
RESULTS: A total of 47 studies were included. Evaluation practices were highly heterogeneous. Expert evaluation (n = 25, 53%) and reference-based evaluations were most common (n = 25, 53%), followed by no-reference metrics (n = 24, 51%), and task-based evaluations (n = 24, 51%). The most commonly used individual metrics were peak signal-to-noise ratio (PSNR) (n = 16, 34%), structural similarity index (SSIM) (n = 15, 32%), mean absolute error (MAE) (n = 12, 26%), and Fréchet Inception Distance (FID) (n = 12, 26%).
CONCLUSION: Evaluation strategies for synthetic medical imaging showed substantial variability and no single metric captured fidelity, realism, diversity, and clinical validity simultaneously. Metric choice is often dictated by data availability rather than clinical purpose. A task-specific, layered evaluation framework could improve comparability and facilitate clinical adoption.
引用
- de Wilde, Schärli, Ackermann, et al. Evaluation metrics for synthetic medical imaging. Eur J Radiol. 2026. PMID: 42659896. doi:10.1016/j.ejrad.2026.113124
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。