核医学与人工智能：评估的最佳实践（RELAINCE 指南）。
Nuclear Medicine and Artificial Intelligence: Best Practices for Evaluation (the RELAINCE Guidelines).
机构信息
Department of Biomedical Engineering and Mallinckrodt Institute of Radiology, Washington University in St. Louis, Missouri;
Department of Radiology, University of Wisconsin-Madison, Madison, Wisconsin.
出版信息
J Nucl Med. 2022 Sep;63(9):1288-1299. doi: 10.2967/jnumed.121.263239. Epub 2022 May 26.
Abstract
An important need exists for strategies to perform rigorous objective clinical-task-based evaluation of artificial intelligence (AI) algorithms for nuclear medicine. To address this need, we propose a 4-class framework to evaluate AI algorithms for promise, technical task-specific efficacy, clinical decision making, and postdeployment efficacy. We provide best practices to evaluate AI algorithms for each of these classes. Each class of evaluation yields a claim that provides a descriptive performance of the AI algorithm. Key best practices are tabulated as the RELAINCE (Recommendations for EvaLuation of AI for NuClear medicinE) guidelines. The report was prepared by the Society of Nuclear Medicine and Molecular Imaging AI Task Force Evaluation team, which consisted of nuclear-medicine physicians, physicists, computational imaging scientists, and representatives from industry and regulatory agencies.
摘要
对于人工智能（AI）算法在核医学中的严格客观临床任务评估策略存在重要需求。为满足这一需求，我们提出了一个 4 级框架，用于评估 AI 算法的承诺、特定技术任务的功效、临床决策和部署后的功效。我们为每一类评估提供了最佳实践。每一类评估都会产生一个声明，提供对 AI 算法的描述性性能。关键最佳实践被列在 RELAINCE（核医学中 AI 评估的建议）指南中。该报告由核医学和分子成像 AI 工作组评估团队编写，成员包括核医学医师、物理学家、计算成像科学家以及来自工业界和监管机构的代表。
相似文献
1
Nuclear Medicine and Artificial Intelligence: Best Practices for Evaluation (the RELAINCE Guidelines).核医学与人工智能：评估的最佳实践（RELAINCE 指南）。
J Nucl Med. 2022 Sep;63(9):1288-1299. doi: 10.2967/jnumed.121.263239. Epub 2022 May 26.
2
Nuclear Medicine and Artificial Intelligence: Best Practices for Algorithm Development.核医学与人工智能：算法开发的最佳实践。
J Nucl Med. 2022 Apr;63(4):500-510. doi: 10.2967/jnumed.121.262567. Epub 2021 Nov 5.
3
Artificial Intelligence in Nuclear Medicine: Opportunities, Challenges, and Responsibilities Toward a Trustworthy Ecosystem.人工智能在核医学中的应用：迈向值得信赖的生态系统的机遇、挑战和责任。
J Nucl Med. 2023 Feb;64(2):188-196. doi: 10.2967/jnumed.121.263703. Epub 2022 Dec 15.
4
AAPM task group report 273: Recommendations on best practices for AI and machine learning for computer-aided diagnosis in medical imaging.AAPM 工作组报告 273：关于医学影像计算机辅助诊断中人工智能和机器学习的最佳实践建议。
Med Phys. 2023 Feb;50(2):e1-e24. doi: 10.1002/mp.16188. Epub 2023 Jan 6.
5
Framework and metrics for the clinical use and implementation of artificial intelligence algorithms into endoscopy practice: recommendations from the American Society for Gastrointestinal Endoscopy Artificial Intelligence Task Force.将人工智能算法临床应用及实施于内镜检查实践的框架与指标：美国胃肠内镜学会人工智能特别工作组的建议
Gastrointest Endosc. 2023 May;97(5):815-824.e1. doi: 10.1016/j.gie.2022.10.016. Epub 2023 Feb 8.
6
Artificial intelligence for nuclear medicine in oncology.人工智能在肿瘤核医学中的应用。
Ann Nucl Med. 2022 Feb;36(2):123-132. doi: 10.1007/s12149-021-01693-6. Epub 2022 Jan 14.
7
Ethical Considerations for Artificial Intelligence in Medical Imaging: Deployment and Governance.人工智能在医学影像中的伦理考量：部署与治理。
J Nucl Med. 2023 Oct;64(10):1509-1515. doi: 10.2967/jnumed.123.266110. Epub 2023 Aug 24.
8
The Low Rate of Adherence to Checklist for Artificial Intelligence in Medical Imaging Criteria Among Published Prostate MRI Artificial Intelligence Algorithms.发表的前列腺 MRI 人工智能算法中，对医学影像学人工智能检查表的依从率低。
J Am Coll Radiol. 2023 Feb;20(2):134-145. doi: 10.1016/j.jacr.2022.05.022. Epub 2022 Jul 31.
9
An EANM position paper on the application of artificial intelligence in nuclear medicine.核医学中人工智能应用的 EANM 立场文件。
Eur J Nucl Med Mol Imaging. 2022 Dec;50(1):61-66. doi: 10.1007/s00259-022-05947-x. Epub 2022 Aug 25.
10
Artificial Intelligence in Nuclear Medicine.人工智能在核医学中的应用
J Nucl Med. 2019 Sep;60(Suppl 2):29S-37S. doi: 10.2967/jnumed.118.220590.
引用本文的文献
1
Emerging Diagnostic Approaches for Musculoskeletal Disorders: Advances in Imaging, Biomarkers, and Clinical Assessment.肌肉骨骼疾病的新兴诊断方法：影像学、生物标志物及临床评估的进展
Diagnostics (Basel). 2025 Jun 27;15(13):1648. doi: 10.3390/diagnostics15131648.
2
Summary Report of the SNMMI AI Task Force Radiomics Challenge 2024.2024年SNMMI人工智能特别工作组放射组学挑战赛总结报告。
J Nucl Med. 2025 Aug 1;66(8):1169-1175. doi: 10.2967/jnumed.124.269425.
3
Characterization of artificial intelligence performance for lesion detection using synthetic lesions in PET imaging.利用PET成像中的合成病变对人工智能病变检测性能进行表征。
Med Phys. 2025 Jun;52(6):3994-4007. doi: 10.1002/mp.17694. Epub 2025 Feb 24.
4
Artificial Intelligence-Enhanced Perfusion Scoring Improves the Diagnostic Accuracy of Myocardial Perfusion Imaging.人工智能增强灌注评分提高了心肌灌注成像的诊断准确性。
J Nucl Med. 2025 Apr 1;66(4):648-653. doi: 10.2967/jnumed.124.268079.
5
Optimizing Cancer Treatment: Exploring the Role of AI in Radioimmunotherapy.优化癌症治疗：探索人工智能在放射免疫治疗中的作用。
Diagnostics (Basel). 2025 Feb 6;15(3):397. doi: 10.3390/diagnostics15030397.
6
AI for IMPACTS Framework for Evaluating the Long-Term Real-World Impacts of AI-Powered Clinician Tools: Systematic Review and Narrative Synthesis.用于评估人工智能驱动的临床医生工具长期现实世界影响的AI for IMPACTS框架：系统评价与叙述性综合分析
J Med Internet Res. 2025 Feb 5;27:e67485. doi: 10.2196/67485.
7
Technological Advances in SPECT and SPECT/CT Imaging.单光子发射计算机断层扫描（SPECT）及SPECT/CT成像技术的进展
Diagnostics (Basel). 2024 Jul 4;14(13):1431. doi: 10.3390/diagnostics14131431.
8
Inspiring a convergent engineering approach to measure and model the tissue microenvironment.激发一种融合工程方法来测量和模拟组织微环境。
Heliyon. 2024 Jun 8;10(12):e32546. doi: 10.1016/j.heliyon.2024.e32546. eCollection 2024 Jun 30.
9
Pitfalls in Developing Machine Learning Models for Predicting Cardiovascular Diseases: Challenge and Solutions.机器学习模型在预测心血管疾病中的陷阱：挑战与解决方案。
J Med Internet Res. 2024 Jul 26;26:e47645. doi: 10.2196/47645.
10
Is Automatic Tumor Segmentation on Whole-Body F-FDG PET Images a Clinical Reality?全身 F-FDG PET 图像的自动肿瘤分割是否成为临床现实？
J Nucl Med. 2024 Jul 1;65(7):995-997. doi: 10.2967/jnumed.123.267183.
本文引用的文献
1
A Projection-Domain Low-Count Quantitative SPECT Method for -Particle-Emitting Radiopharmaceutical Therapy.一种用于发射β粒子放射性药物治疗的投影域低计数定量单光子发射计算机断层扫描方法。
IEEE Trans Radiat Plasma Med Sci. 2023 Jan;7(1):62-74. doi: 10.1109/trpms.2022.3175435. Epub 2022 May 23.
2
Investigating the limited performance of a deep-learning-based SPECT denoising approach: An observer-study-based characterization.基于深度学习的单光子发射计算机断层扫描去噪方法的性能局限性研究：基于观察者研究的特征描述。
Proc SPIE Int Soc Opt Eng. 2022 Feb-Mar;12035. doi: 10.1117/12.2613134. Epub 2022 Apr 4.
3
A Hybrid Approach for Approximating the Ideal Observer for Joint Signal Detection and Estimation Tasks by Use of Supervised Learning and Markov-Chain Monte Carlo Methods.基于监督学习和马尔可夫链蒙特卡罗方法的联合信号检测与估计任务中理想观察者的混合逼近方法。
IEEE Trans Med Imaging. 2022 May;41(5):1114-1124. doi: 10.1109/TMI.2021.3135147. Epub 2022 May 2.
4
Nuclear Medicine and Artificial Intelligence: Best Practices for Algorithm Development.核医学与人工智能：算法开发的最佳实践。
J Nucl Med. 2022 Apr;63(4):500-510. doi: 10.2967/jnumed.121.262567. Epub 2021 Nov 5.
5
A physics and learning-based transmission-less attenuation compensation method for SPECT.一种基于物理和学习的单光子发射计算机断层扫描无传输衰减补偿方法。
Proc SPIE Int Soc Opt Eng. 2021 Feb;11595. doi: 10.1117/12.2582350. Epub 2021 Feb 15.
6
Objective Task-Based Evaluation of Artificial Intelligence-Based Medical Imaging Methods:: Framework, Strategies, and Role of the Physician.基于目标的人工智能医学成像方法的客观评估：框架、策略和医师的作用。
PET Clin. 2021 Oct;16(4):493-511. doi: 10.1016/j.cpet.2021.06.013.
7
The Clinician and Dataset Shift in Artificial Intelligence.临床医生与人工智能中的数据集偏移
N Engl J Med. 2021 Jul 15;385(3):283-286. doi: 10.1056/NEJMc2104626.
8
A Bayesian approach to tissue-fraction estimation for oncological PET segmentation.基于贝叶斯方法的肿瘤 PET 分割组织分数估计。
Phys Med Biol. 2021 Jun 14;66(12). doi: 10.1088/1361-6560/ac01f4.
9
CT-less Direct Correction of Attenuation and Scatter in the Image Space Using Deep Learning for Whole-Body FDG PET: Potential Benefits and Pitfalls.使用深度学习对全身 FDG PET 在图像空间中进行无 CT 的衰减和散射直接校正：潜在益处与陷阱
Radiol Artif Intell. 2020 Dec 2;3(2):e200137. doi: 10.1148/ryai.2020200137. eCollection 2021 Mar.
10
Checklist for Artificial Intelligence in Medical Imaging (CLAIM): A Guide for Authors and Reviewers.医学影像人工智能清单（CLAIM）：作者和审稿人指南
Radiol Artif Intell. 2020 Mar 25;2(2):e200029. doi: 10.1148/ryai.2020200029. eCollection 2020 Mar.
Zotero 插件