Advances in Clinical Medicine
Vol. 14  No. 01 ( 2024 ), Article ID: 80014 , 7 pages
10.12677/ACM.2024.141215

人工智能在PSMA PET/CT中的应用

张皓哲1,2*,曹敏2*,冀明1,刘洪年1#

1山东第一医科大学第二附属医院泌尿外科,山东 泰安

2山东第一医科大学(山东省医学科学院)研究生院,山东 济南

收稿日期:2023年12月25日;录用日期:2024年1月19日;发布日期:2024年1月29日

摘要

前列腺特异性膜抗原(Prostate-Specific Membrane Antigen, PSMA)正电子发射断层扫描(Positron Emission Tomography, PET)/计算机断层扫描(Computed Tomography, CT)已成为前列腺癌重要的成像技术。随着研究和应用的深入拓展,人工智能(Artificial Intelligence, AI)开始应用于PSMA PET/CT。本文分析了人工智能在PSMA PET/CT前列腺癌成像中的发展和应用。然后综合介绍了目前AI技术在前列腺病灶检测、分类、分期、治疗及预后等方面的应用现状。大量研究已证明,AI技术在PSMA PET/CT方面取得了显著的成果,但仍面临一些挑战。未来,AI技术有望为医生提供更准确和个体化的前列腺癌诊断和治疗决策支持。

关键词

人工智能,前列腺特异性膜抗原,正电子发射体层成像,前列腺癌

The Application of Artificial Intelligence in PSMA PET/CT

Haozhe Zhang1,2*, Min Cao2*, Ming Ji1, Hongnian Liu1#

1Department of Urology, The Second Affiliated Hospital of Shandong First Medical University, Tai’an Shandong

2Graduate School of Shandong First Medical University (Shandong Academy of Medical Science), Jinan Shandong

Received: Dec. 25th, 2023; accepted: Jan. 19th, 2024; published: Jan. 29th, 2024

ABSTRACT

Prostate-specific membrane antigen (PSMA) positron emission tomography/computed tomography (PET/CT) has emerged as an essential imaging modality for prostate cancer. With the advancement and application of research, the integration of artificial intelligence (AI) into PSMA PET/CT imaging has begun. This article critically examines the development and utilization of AI in PSMA PET/CT imaging for prostate cancer. Additionally, it provides a comprehensive overview of the current application status of AI technology in detecting, classifying, staging, treating, and prognosticating prostate lesions. A number of investigations have showcased the remarkable achievements of AI technology in the realm of PSMA PET/CT, albeit still encountering certain challenges. In the future, AI technology is expected to facilitate more precise and individualized support for the diagnosis and treatment decision-making of prostate cancer.

Keywords:Artificial Intelligence, Prostate-Specific Membrane Antigen, Positron Emission Tomography, Prostate Cancer

Copyright © 2024 by author(s) and Hans Publishers Inc.

This work is licensed under the Creative Commons Attribution International License (CC BY 4.0).

http://creativecommons.org/licenses/by/4.0/

1. 引言

前列腺癌(Prostate Cancer, PCa)在全球男性肿瘤中发病率排名第二,死亡率排名第五 [1] 。中国发病率和死亡率逐年上升 [2] 。PCa的治疗中,癌症的分期对于治疗及预后具有重要影响 [3] 。对于晚期PCa患者,衡量肿瘤代谢负荷随时间的变化对于指导治疗和评估治疗效果至关重要。由于缺乏准确反映疾病传播的血液标志物,需要依赖成像技术评估 [4] 。近年来,前列腺特异性膜抗原(PSMA)正电子发射断层扫描/计算机断层扫描(PET/CT)在成像中愈发重要,与传统成像相比,PSMA PET/CT的检测效果更好,对高风险原发性的分期有重要作用,在近期的一项研究中,PSMA PET-CT的准确度比传统成像高23% [5] 。根治性治疗对前列腺癌(PCa)的有效性通常取决于准确的疾病分期,包括确定前列腺外病灶和盆腔淋巴结的转移性存在。因此,在选择有明确治疗意图的患者时,使用PSMA成像来确定疾病范围并在放射治疗前仔细选择患者非常重要。随着机器学习(Machine Learning, ML)和深度学习(Deep Learning, DL)在医学领域的普及和进步,人工智能(Artificial Intelligence, AI)可以在PSMA PET/CT图像解释、分析和辅助诊断过程中提供有价值的帮助。本综述就人工智能在PSMAPET/CT中的发展和应用进行综述。

2. PSMA PET/CT在PCa中的应用

PSMA是一种II型跨膜糖蛋白,在PCa中的表达水平是其他组织(如良性前列腺、肾脏等)的100~1000倍,是重要的临床生物标志物 [6] 。PSMA PET/CT在预测前列腺癌预后方面发挥了重要作用 [7] 。通过分析PSMA PET/CT图像,可以获取有关肿瘤活性和代谢情况的信息,这对于评估肿瘤的侵袭性和预后具有重要意义,在Klingenberg等人 [5] 的实验中,68Ga-PSMA PET/CT检测肿瘤分期的敏感性、特异性、阳性和阴性预测值以及准确性分别为30.6%、96.5%、68.8%、84.5%和83.1%。研究表明,PSMA PET/CT能够提供关于前列腺癌的更准确的分期和分级信息,因为它可以显示肿瘤的大小、位置、淋巴结受累情况和可能的远处转移 [8] 。这些因素都与患者的预后密切相关 [9] 。此外,PSMA PET/CT还可以用于评估治疗前后的疾病进展和反应。对于治疗后复发的患者,PSMA PET/CT可以帮助确定复发灶的位置、活跃度和扩散情况,从而指导后续治疗策略的选择 [10] 。国际上已有多项研究表明,对比传统成像,PSMA PET/CT对PCa的初诊有更高的诊断价值。在Eiber等人 [11] 的研究中,对于53名PCa患者,前列腺多参数磁共振成像(multiparametric magnetic resonance imaging, mpMRI)和68Ga-PSMA PET/CT的检出率分别为66%和92%。此外,PSMA PET/CT在识别淋巴结疾病 [12] 和骨转移 [13] 方面优于传统成像。与传统影像工具不同,PSMA PET/CT被证明适用于检测复发。Hoffmann等人 [14] 的研究中18F-PSMA-1007和68Ga-PSMA-11 PET/CT生化复发(Biochemical recurrence, BCR)的患者均有较高的检出率。通过PSMA PET/CT成像更早检测到复发可促使临床医生更早地考虑治疗。PSMA PET/CT结果可以指导临床决策、治疗选择或全身姑息治疗的决策。

3. 人工智能在医学领域的应用

人工智能在医学领域的应用取得了许多重要的突破。AI可以应用于医学影像,如X射线、CT扫描、MRI等。通过ML和DL模型,可以实现高准确度的图像识别和分析,用于皮肤病、癌症、眼病等疾病的辅助诊断 [15] 。AI还可以分析患者的医疗数据,提供疾病的预测和风险评估 [16] 。此外,AI可以实现自动化医学影像的分析和报告生成 [17] 。AI可以驱动医疗机器人的发展,用于辅助手术和康复治疗。机器人手术系统可以精确操作,减少手术风险和伤害 [18] 。同时,机器人还可以通过人工智能算法提供实时的术中导航和辅助决策。

4. 人工智能在PSMA PET/CT中的应用

人工智能可用于自动诊断和解读PSMA PET/CT图像。通过训练神经网络模型,AI可以识别和标记出图像中的区域,提供快速准确的诊断结果。这种自动化分析可以缩短诊断时间,提高诊断准确性,并且能够处理大量的图像数据。

4.1. 前列腺病灶的识别与分割

许多研究提出应用人工智能自动检测和分割原发性肿瘤和转移性病灶的方法。与传统的mpMRI相比,PSMA PET/CT在检测前列腺内病变方面具有更高的灵敏度和特异性 [19] 。Yi等人 [20] 同样基于68Ga-PSMA-11 PET构建了随机森林模型,该模型能够准确预测在68Ga-PSMA-11 PET上不可见的前列腺内病变(受试者操作特征曲线下面积AUC,0.903)。在另一项研究 [21] ,Ricarda等人采用68Ga-PSMA-11 PET/CT训练的模型,可用于检测CT中不可见的PCa骨转移病灶。

Zhao等人 [22] 开发了一种深度学习方法,用于在68Ga-PSMA-11 PET/CT图像上自动检测和分割转移性去势抵抗前列腺癌(mCRPC)病变。与手动分割结果相比,该方法在骨病变和淋巴结的检测方面表现出较高的敏感性(分别为99%和90%),但在前列腺病变的检测方面敏感性较低(61%)。在其验证队列中具有类似的准确性。Hartenstein等人 [23] 基于68GA-PSMA-11 PET/CT构建的卷积神经网络(Convolutional Neural Networks, CNN)模型在盆腔淋巴结转移的检测方面的AUC (0.95)高于影像科医师的AUC (0.81),证明CNN有潜力建立一个性能良好的基于PSMAPET/CT的PCa淋巴结转移生物标志物。Holzschuh等人 [24] 的研究结果表明CNN可准确描绘PSMA-PET中的前列腺内肿瘤总体积(Gross Tumor Volume, GTV),其准确率与影像科医师相当,CNN预测每位患者平均需要3.81秒较影像科医师更快,可节省大量时间。

Simon等人 [25] 基于68Ga-PSMA-11 PET/CT构建的影像组学模型可较准确地预测PCa患者放疗后生化复发(biochemical recurrence, BCR),Kendrick [26] 开发的CNN模型,用于对193名生化复发性前列腺癌患者进行68Ga-PSMA-11 PET/CT进行全自动全身病变分割。该模型的灵敏度为73%,阳性预测值为88%,假阳性预测率较低。

Cysouw等人 [27] 开发了一个影像组学模型,用于检测76名患有中高危前列腺癌并计划进行根治性前列腺切除术的患者的术前18F-DCFPyL-PET/CT中的转移性病灶。该模型在淋巴结和远处转移的检测方面表现出良好的区分能力(AUC, 0.86)。研究结论显示,该模型可以无创地预测低风险患者,从而避免扩大盆腔淋巴结清扫术。然而,该研究缺乏多中心验证。

4.2. 前列腺病灶的分类

除前列腺组织外,PSMA在其他组织器官中也有表达 [27] 。一些研究表明,PSMA在正常组织中的表达包括小肠上皮细胞、肾小管上皮细胞、脑组织和肝脏等。这意味着PSMA PET扫描或其他PSMA相关诊断技术在评估前列腺癌时,可能会面临特异性的挑战,利用AI的方法进行PCa分类研究能够显著提升分类精度。

Zang等人 [28] 基于68Ga-PSMA-11 PET/CT构建的影像组学模型可用于预测PCa。该模型的诊断效能(AUC, 0.85)优于影像科医师(AUC, 0.63),且灵敏度较高。Erle等人 [29] 通过提取的影像组学特征构建的三种ML分类模型,在鉴别68Ga-PSMA-11 PET/CT图像中的生理性摄取和恶性摄取方面,均有不错的特异性。Leung等人 [30] 的研究在开发的影像组学模型可对PCa患者的18F-DCFPyL-PET/CT图像进行PSMA-RADS和PCa分类,测试集AUC可达0.92。

4.3. 前列腺癌的治疗与预后

PSMA PET/CT在前列腺癌的预后预测中扮演着重要的角色 [31] 。通过分析PSMA PET/CT的定量参数,如最大标准摄取值(SUVmax)和病灶密度等,可以预测前列腺癌患者的预后 [32] 。高SUVmax和更高的病灶密度通常与较差的预后相关 [33] 。PCa对放疗敏感,因此放疗是作为一线治疗选择。177Lu因具有较高的安全性,并且具备显像和治疗的潜力,被广泛应用于PSMA阳性的晚期或转移性前列腺癌患者的治疗中,通过放射性核素的投放,实现对肿瘤细胞的辐射治疗,帮助控制疾病进展并改善患者的预后 [34] 。

Capobianco等人 [35] 基于68Ga-PSMA-11 PET/CT开发的深度学习方法模型可有效评估癌症分期和肿瘤负荷。Samuele等人 [36] 的研究中,其基于68Ga-PSMA-11 PET影像组学可准确无创地预测ISUP级。Kendrick等人 [26] 的研究中,基于68Ga-PSMA PET/CT和CNN的模型与总体生存率显著相关的结果。可用作与患者总生存期相关的潜在的预后生物标志物。Yao等人 [37] 在18F-PSMA PET/CT上使用基于支持向量机(SVM)的影像组学模型来预测Gleason评分、囊外扩张和血管侵犯,并评估了分割过程的不同阈值的SUVmax如何影响预测准确性。在另外两项研究中 [38] ,对于接受全身治疗的转移性疾病患者,包括ADT、化疗和177Lu-PSMA,被证实治疗前PSMA PET提取的放射学特征可用于预测转移性PCa患者对全身治疗的反应。Matthijs等人 [27] 确定了基于机器学习的定量分析18F-DCFPyL-PET/CT指标预测转移性疾病或高危病理肿瘤特征的能力。

Moazemi等人 [39] 的研究证明,基于68Ga-PSMA PET/CT影像组学特征的机器学习模型有望预测对177Lu-PSMA治疗的反应,AUC为0.80,敏感性和特异性均为75%。在Moazemi [40] 的另一项研究中,基于68Ga-PSMA-PET/CT的影像组学特征以及患者特异性临床参数构建的风险模型,可预测接受177Lu-PSMA治疗的晚期PCa患者的总生存期(Overall Survival, OS)。Kelk等人 [41] 构建了一个基于体素的剂量学和病灶内影像组学的剂量学分析模型,可评估放射治疗计划在病灶区域内的输送剂量。利用基于体素的剂量学分析模型,可以获得更准确和可靠的剂量学信息。使用病灶内影像组学,可将剂量学数据与病灶的形态特征和影像特征相结合,从而更全面地评估放射治疗计划的效果。这一研究成果在常规剂量学中具有实用性,能够为放射治疗医师提供更好的决策依据,还可以在放射治疗计划评估过程中用于优化剂量分布,提高治疗效果,减少对正常组织的损伤。

5. 展望

大量的研究证明,人工智能方法的快速发展,特别是机器学习和深度学习方法,为PSMA PET/CT在PCa的诊断性能提供巨大的潜力。可以提高前列腺的定位和分割准确性,实现自动化的病灶检测和定量分析,进行预后评估和治疗响应监测,并通过多模态图像融合提高诊断性能。现有的人工分割方法需要耗费大量的时间和精力。而且在解读过程中存在着解读者之间的主观差异。借助人工智能工具在PSMA PET/CT解读中的应用,可以支持放射科医生和核医学医生提高解读准确性,减少解读者间的差异,节省报告时间,并为临床医生提供定量数据,进一步提升PSMA PET/CT的应用效果。

该领域的研究也存在一定挑战。目前大多数都是单一中心及回顾性研究,且缺乏大样本的数据,结果可能具有一定的局限性。由于数据集的局限性,模型的泛化能力可能受到限制,无法很好地应用于其他数据集。因此,在推广和应用这些模型时,需要谨慎解读结果,并在实际应用中进行验证和调整。进一步的研究可以通过多中心合作,前瞻性实验、增加样本数量、覆盖更广泛的地区和人群,以及使用更多源自不同数据集的数据来提高模型的泛化性能和应用范围。目前,在图像重建领域存在多种不同的成像模式和协议,由于扫描参数的差异会对特征的提取产生影响,使其具有可变性。实现统一的扫描参数和成像协议是未来发展的方向之一。通过制定标准、共享数据和开发智能算法,可以提高图像重建的可比性和准确性,促进成像技术的进一步发展。目前PSMA PET/CT在多组学研究方面的应用相对较少,将PSMA PET/CT与基因组学 [42] 、病理组学 [43] 等方法的联合应用,将有助于更全面地了解前列腺癌的生物学特征和发展机制,为个体化治疗和预后评估提供更准确的信息。

6. 小结

人工智能技术在前列腺癌PSMA PET/CT分析和解释方面的挑战具有巨大的潜力。人工智能在PSMA PET/CT自动分割和定位、病灶检测和分类、数据集成和分析和预后评估和治疗响应监测等方面都有突出的表现。随着更多的研究和应用,人工智能技术有望为医生提供更准确、快速和个体化的前列腺癌诊断和治疗决策支持。

文章引用

张皓哲,曹 敏,冀 明,刘洪年. 人工智能在PSMA PET/CT中的应用
The Application of Artificial Intelligence in PSMA PET/CT[J]. 临床医学进展, 2024, 14(01): 1501-1507. https://doi.org/10.12677/ACM.2024.141215

参考文献

  1. 1. Sung, H., Ferlay, J., Siegel, R.L., et al. (2021) Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: A Cancer Journal for Clinicians, 71, 209-249. https://doi.org/10.3322/caac.21660

  2. 2. Qiu, H., Cao, S. and Xu, R. (2021) Cancer Incidence, Mortality, and Bur-den in China: A Time-Trend Analysis and Comparison with the United States and United Kingdom Based on the Global Epidemiological Data Released in 2020. Cancer Communications (London), 41, 1037-1048. https://doi.org/10.1002/cac2.12197

  3. 3. Parker, C., Castro, E., Fizazi, K., et al. (2020) Prostate Cancer: ESMO Clinical Practice Guidelines for Diagnosis, Treatment and Follow-Up. Annals of Oncology, 31, 1119-1134. https://doi.org/10.1016/j.annonc.2020.06.011

  4. 4. Kuten, J., Kesler, M. and Even-Sapir, E. (2021) The Role of PSMA PET/CT in Imaging Prostate Cancer. Harefuah, 160, 455-461.

  5. 5. Klingenberg, S., Jochumsen, M.R., Ulhøi, B.P., et al. (2021) 68Ga-PSMA PET/CT for Primary Lymph Node and Distant Metastasis NM Staging of High-Risk Prostate Cancer. Journal of Nuclear Medicine, 62, 214-220. https://doi.org/10.2967/jnumed.120.245605

  6. 6. Maurer, T., Eiber, M., Schwaiger, M., et al. (2016) Current Use of PSMA-PET in Prostate Cancer Management. Nature Reviews Urology, 13, 226-235. https://doi.org/10.1038/nrurol.2016.26

  7. 7. Bagguley, D., Ong, S., Buteau, J.P., et al. (2021) Role of PSMA PET/CT Imaging in the Diagnosis, Staging and Restaging of Prostate Cancer. Future Oncology, 17, 2225-2241. https://doi.org/10.2217/fon-2020-1293

  8. 8. Barbosa, Á.R.G., Amaral, B.S., Lourenço, D.B., et al. (2022) Accura-cy of 68Ga-PSMA PET-CT and PET-MRI in Lymph Node Staging for Localized Prostate Cancer. Einstein (Sao Paulo), 20, eAO6599. https://doi.org/10.31744/einstein_journal/2022AO6599

  9. 9. Pozdnyakov, A., Kulanthaivelu, R., Bauman, G., et al. (2023) The Impact of PSMA PET on the Treatment and Outcomes of Men with Biochemical Recurrence of Prostate Cancer: A Systematic Review and Meta-Analysis. Prostate Cancer and Prostatic Diseases, 26, 240-248. https://doi.org/10.1038/s41391-022-00544-3

  10. 10. Rüschoff, J.H., Ferraro, D.A., Muehlematter, U.J., et al. (2021) What’s behind 68Ga-PSMA-11 Uptake in Primary Prostate Cancer PET? Investigation of Histopathological Parameters and Immunohistochemical PSMA Expression Patterns. European Journal of Nuclear Medicine and Molecular Imaging, 48, 4042-4053. https://doi.org/10.1007/s00259-021-05501-1

  11. 11. Eiber, M., Weirich, G., Holzapfel, K., et al. (2016) Simultaneous 68Ga-PSMA HBED-CC PET/MRI Improves the Localization of Primary Prostate Cancer. European Urology, 70, 829-836. https://doi.org/10.1016/j.eururo.2015.12.053

  12. 12. Peng, L., Li, J., Meng, C., et al. (2020) Can 68Ga-Prostate Specific Membrane Antigen Positron Emission Tomography/Computerized Tomography Provide an Ac-curate Lymph Node Staging for Patients with Medium/High Risk Prostate Cancer? A Diagnostic Meta-Analysis. Radia-tion Oncology, 15, Article No. 227. https://doi.org/10.1186/s13014-020-01675-4

  13. 13. Qiu, D.X., Li, J., Zhang, J.W., et al. (2022) Dual-Tracer PET/CT-Targeted, mpMRI-Targeted, Systematic Biopsy, and Combined Biopsy for the Diagnosis of Prostate Cancer: A Pilot Study. European Journal of Nuclear Medicine and Molecular Imaging, 49, 2821-2832. https://doi.org/10.1007/s00259-021-05636-1

  14. 14. Hoffmann, M.A., von Eyben, F.E., Fischer, N., et al. (2022) Comparison of [18F]PSMA-1007 with [68Ga]Ga-PSMA-11 PET/CT in Restaging of Prostate Cancer Patients with PSA Relapse. Cancers (Basel), 14, Article No. 1479. https://doi.org/10.3390/cancers14061479

  15. 15. Seetharaman, A., Bhattacharya, I., Chen, L.C., et al. (2021) Auto-mated Detection of Aggressive and Indolent Prostate Cancer on Magnetic Resonance Imaging. Medical Physics, 48, 2960-2972. https://doi.org/10.1002/mp.14855

  16. 16. Xie, W., Reder, N.P., Koyuncu, C., et al. (2022) Prostate Can-cer Risk Stratification via Nondestructive 3D Pathology with Deep Learning-Assisted Gland Analysis. Cancer Research, 82, 334-345. https://doi.org/10.1158/0008-5472.CAN-21-2843

  17. 17. Vergani, V., Razavi, R., Puyol-Antón, E., et al. (2021) Deep Learning for Classification and Selection of Cine CMR Images to Achieve Fully Automated Quality-Controlled CMR Analysis from Scanner to Report. Frontiers in Cardiovascular Medicine, 8, Article ID: 742640. https://doi.org/10.3389/fcvm.2021.742640

  18. 18. Bravi, C.A., Paciotti, M., Sarchi, L., et al. (2022) Robot-Assisted Radical Prostatectomy with the Novel Hugo Robotic System: Initial Experience and Optimal Surgical Set-Up at a Tertiary Referral Robotic Center. European Urology, 82, 233-237. https://doi.org/10.1016/j.eururo.2022.04.029

  19. 19. Spohn, S.K.B., Kramer, M., Kiefer, S., et al. (2020) Comparison of Manual and Semi-Automatic [18F]PSMA-1007 PET Based Contouring Techniques for Intraprostatic Tumor Delineation in Patients with Primary Prostate Cancer and Validation with Histopathology as Standard of Reference. Frontiers in Oncology, 10, Article ID: 600690. https://doi.org/10.3389/fonc.2020.600690

  20. 20. Yi, Z., Hu, S., Lin, X., et al. (2022) Machine Learning-Based Pre-diction of Invisible Intraprostatic Prostate Cancer Lesions on 68Ga-PSMA-11 PET/CT in Patients with Primary Prostate Cancer. European Journal of Nuclear Medicine and Molecular Imaging, 49, 1523-1534. https://doi.org/10.1007/s00259-021-05631-6

  21. 21. Hinzpeter, R., Baumann, L., Guggenberger, R., et al. (2022) Ra-diomics for Detecting Prostate Cancer Bone Metastases Invisible in CT: A Proof-of-Concept Study. European Radiology, 32, 1823-1832. https://doi.org/10.1007/s00330-021-08245-6

  22. 22. Zhao, Y., Gafita, A., Vollnberg, B., et al. (2020) Deep Neural Network for Automatic Characterization of Lesions on 68Ga-PSMA-11 PET/CT. European Journal of Nuclear Medicine and Molecular Imaging, 47, 603-613. https://doi.org/10.1007/s00259-019-04606-y

  23. 23. Hartenstein, A., Lübbe, F., Baur, A.D.J., et al. (2020) Prostate Cancer Nodal Staging: Using Deep Learning to Predict 68Ga-PSMA-Positivity from CT Imaging Alone. Scientific Re-ports, 10, Article No. 3398. https://doi.org/10.1038/s41598-020-60311-z

  24. 24. Holzschuh, J.C., Mix, M., Ruf, J., et al. (2023) Deep Learning based Automated Delineation of the Intraprostatic Gross Tumour Volume in PSMA-PET for Patients with Primary Prostate Cancer. Radiotherapy and Oncology, 2023, Article ID: 109774. https://doi.org/10.1016/j.radonc.2023.109774

  25. 25. Spohn, S.K.B., Schmidt-Hegemann, N.S., Ruf, J., et al. (2023) Development of PSMA-PET-Guided CT-Based Radiomic Signature to Predict Biochemical Recurrence after Salvage Ra-diotherapy. European Journal of Nuclear Medicine and Molecular Imaging, 50, 2537-2547. https://doi.org/10.1007/s00259-023-06195-3

  26. 26. Kendrick, J., Francis, R.J., Hassan, G.M., et al. (2022) Fully Automatic Prognostic Biomarker Extraction from Metastatic Prostate Lesion Segmentations in Whole-Body [68Ga]Ga-PSMA-11 PET/CT Images. European Journal of Nuclear Medicine and Molecular Imaging, 50, 67-79. https://doi.org/10.1007/s00259-022-05927-1

  27. 27. Cysouw, M.C.F., Jansen, B.H.E., van de Brug, T., et al. (2021) Machine Learning-Based Analysis of [18F]DCFPyL PET Radiomics for Risk Stratification in Primary Prostate Cancer. European Journal of Nuclear Medicine and Molecular Imaging, 48, 340-349. https://doi.org/10.1007/s00259-020-04971-z

  28. 28. Zang, S., Ai, S., Yang, R., et al. (2022) Development and Vali-dation of 68Ga-PSMA-11 PET/CT-Based Radiomics Model to Detect Primary Prostate Cancer. EJNMMI Research, 12, Article No. 63. https://doi.org/10.1186/s13550-022-00936-5

  29. 29. Erle, A., Moazemi, S., Lütje, S., et al. (2021) Evaluating a Ma-chine Learning Tool for the Classification of Pathological Uptake in Whole-Body PSMA-PET-CT Scans. Tomography, 7, 301-312. https://doi.org/10.3390/tomography7030027

  30. 30. Leung, K.H., Rowe, S.P., Leal, J.P., et al. (2022) Deep Learning and Radiomics Framework for PSMA-RADS Classification of Prostate Cancer on PSMA PET. EJNMMI Research, 12, Article No. 76. https://doi.org/10.1186/s13550-022-00948-1

  31. 31. Gafita, A., Rauscher, I., Fendler, W.P., et al. (2022) Measuring Response in Metastatic Castration-Resistant Prostate Cancer Using PSMA PET/CT: Comparison of RECIST 1.1, aPCWG3, aPERCIST, PPP, and RECIP 1.0 Criteria. European Journal of Nuclear Medicine and Molecular Imaging, 49, 4271-4281. https://doi.org/10.1007/s00259-022-05882-x

  32. 32. Ferdinandus, J., Violet, J., Sandhu, S., et al. (2020) Prognostic Biomarkers in Men with Metastatic Castration-Resistant Prostate Cancer Receiving [177Lu]-PSMA-617. European Jour-nal of Nuclear Medicine and Molecular Imaging, 47, 2322-2327. https://doi.org/10.1007/s00259-020-04723-z

  33. 33. Seifert, R., Herrmann, K., Kleesiek, J., et al. (2020) Semiauto-matically Quantified Tumor Volume Using (68)Ga-PSMA-11 PET as a Biomarker for Survival in Patients with Ad-vanced Prostate Cancer. Journal of Nuclear Medicine, 61, 1786-1792. https://doi.org/10.2967/jnumed.120.242057

  34. 34. Sartor, O., de Bono, J., Chi, K.N., et al. (2021) Luteti-um-177-PSMA-617 for Metastatic Castration-Resistant Prostate Cancer. The New England Journal of Medicine, 385, 1091-1103. https://doi.org/10.1056/NEJMoa2107322

  35. 35. Capobianco, N., Sibille, L., Chantadisai, M., et al. (2022) Whole-Body Uptake Classification and Prostate Cancer Staging in 68Ga-PSMA-11 PET/CT Using Dual-Tracer Learning. European Journal of Nuclear Medicine and Molecular Imaging, 49, 517-526. https://doi.org/10.1007/s00259-021-05473-2

  36. 36. Ghezzo, S., Mapelli, P., Bezzi, C., et al. (2023) Role of [68Ga]Ga-PSMA-11 PET Radiomics to Predict Post-Surgical ISUP Grade in Primary Prostate Cancer. European Journal of Nuclear Medicine and Molecular Imaging, 50, 2548-2560. https://doi.org/10.1007/s00259-023-06187-3

  37. 37. Yao, F., Bian, S., Zhu, D., et al. (2022) Machine Learning-Based Radiomics for Multiple Primary Prostate Cancer Biological Characteristics Prediction with 18F-PSMA-1007 PET: Com-parison among Different Volume Segmentation Thresholds. La Radiologia Medica, 127, 1170-1178. https://doi.org/10.1007/s11547-022-01541-1

  38. 38. Tran, V.T., Tu, S.J. and Tseng, J.R. (2022) 68Ga-PSMA-11 PET/CT Features Extracted from Different Radiomic Zones Predict Response to Androgen Deprivation Therapy in Pa-tients with Advanced Prostate Cancer. Cancers (Basel), 14, Article No. 4838. https://doi.org/10.3390/cancers14194838

  39. 39. Moazemi, S., Erle, A., Khurshid, Z., et al. (2021) Decision-Support for Treatment with 177Lu-PSMA: Machine Learning Predicts Response with High Accuracy Based on PSMA-PET/CT and Clinical Parameters. Annals of Translational Medicine, 9, Article No. 818. https://doi.org/10.21037/atm-20-6446

  40. 40. Moazemi, S., Erle, A., Lütje, S., et al. (2021) Estimating the Potential of Radiomics Features and Radiomics Signature from Pretherapeutic PSMA-PET-CT Scans and Clinical Data for Prediction of Overall Survival When Treated with 177Lu-PSMA. Diagnostics (Basel), 11, Article No. 186. https://doi.org/10.3390/diagnostics11020186

  41. 41. Kelk, E., Ruuge, P., Rohtla, K., et al. (2021) Radiomics Analysis for 177Lu-DOTAGA-(l-y)fk(Sub-KuE) Targeted Radioligand Therapy Dosimetry in Metastatic Prostate Cancer—A Model Based on Clinical Example. Life (Basel), 11, Article No. 170. https://doi.org/10.3390/life11020170

  42. 42. Castaldo, R., Cavaliere, C., Soricelli, A., et al. (2021) Radiomic and Ge-nomic Machine Learning Method Performance for Prostate Cancer Diagnosis: Systematic Literature Review. Journal of Medical Internet Research, 23, e22394. https://doi.org/10.2196/22394

  43. 43. Panunzio, A., Tafuri, A., Princiotta, A., et al. (2021) Omics in Urology: An Overview on Concepts, Current Status and Future Perspectives. Urologia, 88, 270-279. https://doi.org/10.1177/03915603211022960

  44. NOTES

    *共第一作者。

    #通讯作者。

期刊菜单