人工智能在前列腺特异性膜抗原正电子发射断层显像中的价值:进展更新
本综述检索Medline、Embase、Web of Science和IEEE Xplore数据库,纳入从建库至2024年10月的英文原创研究,评估人工智能在前列腺癌PSMA PET扫描中的价值。人工智能在自动检测转移灶、解剖分割、预测PSMA诊疗反应、肿瘤负荷分割、改进放疗计划、区分高组织学分级前列腺内癌及预测前列腺外延伸方面显示出潜力。作者认为需更大规模多中心前瞻性研究验证并提高这些人工智能模型的泛化能力。
要点
- 人工智能在PSMA PET中可自动检测转移灶和进行解剖分割。
- 人工智能可预测PSMA诊疗反应、辅助肿瘤负荷分割并改进放疗计划。
- 人工智能可区分更高组织学分级的前列腺内癌并预测前列腺外延伸。
核心结论
- 人工智能在评估前列腺癌PSMA PET扫描方面有潜力,尤其在检测转移、测量肿瘤负荷、检测高级别前列腺内癌和预测治疗结局方面。
- 作者认为需更大规模多中心前瞻性研究验证并提高这些人工智能模型的泛化能力。
研究设计与样本
- 本综述在Medline、Embase、Web of Science和IEEE Xplore数据库进行文献检索,聚焦利用人工智能评估PSMA PET扫描的研究。
- 纳入从建库至2024年10月的英文原创研究,排除病例报告、病例系列、评论和会议论文集。
主要发现
- 人工智能应用在自动检测转移性疾病和PSMA PET扫描解剖分割方面显示出潜力。
- 人工智能能够预测对PSMA诊疗的反应,并辅助肿瘤负荷分割,改进放疗计划。
- 人工智能可区分具有更高组织学分级的前列腺内前列腺癌并预测前列腺外延伸。
局限性
- 作者认为需更大规模多中心前瞻性研究验证并提高这些人工智能模型的泛化能力。
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)。
原文
This review aims to provide an up-to-date overview of the utility of artificial intelligence (AI) in evaluating prostate-specific membrane antigen (PSMA) positron emission tomography (PET) scans for prostate cancer (PCa). A literature review was conducted on the Medline, Embase, Web of Science, and IEEE Xplore databases. The search focused on studies that utilizes AI to evaluate PSMA PET scans. Original English language studies published from inception to October 2024 were included, while case reports, series, commentaries, and conference proceedings were excluded. AI applications show promise in automating the detection of metastatic disease and anatomical segmentation in PSMA PET scans. AI was also able to predict response to PSMA-based theragnostic and aids in tumor burden segmentation, improving radiotherapy planning. AI could also differentiate intraprostatic PCa with higher histological grade and predict extra-prostatic extension. AI has potential in evaluating PSMA PET scans for PCa, particularly in detecting metastasis, measuring tumor burden, detecting high grade intraprostatic cancer, and predicting treatment outcomes. Larger multicenter prospective studies are necessary to validate and enhance the generalizability of these AI models.
引用
- Liu J, Sandhu K, Woon DTS, et al. The Value of Artificial Intelligence in Prostate-Specific Membrane Antigen Positron Emission Tomography: An Update. Semin Nucl Med. 2025. PMID: 39893058. doi:10.1053/j.semnuclmed.2024.12.001
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。