Automatic lesion segmentation in ⁶⁸Ga-PSMA PET/CT and ¹⁷⁷Lu-PSMA SPECT/CT: added value of PET-guided SPECT in a bicentric study.

IF 2.7 3区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

EJNMMI Research Pub Date : 2026-07-01 DOI:10.1186/s13550-026-01468-y

Solène Perret, Léa Albe, Agathe Edet-Sanson, David Tonnelet, Thomas Carlier, Camilla Kohi, Matthieu Barbaud, Romain Modzelewski, Pierre Vera, Laetitia Augusto, Frédéric Di Fiore, Sébastien Hapdey, Clément Bailly, Arnaud Dieudonné, Pierre Decazes

{"title":"Automatic lesion segmentation in ⁶⁸Ga-PSMA PET/CT and ¹⁷⁷Lu-PSMA SPECT/CT: added value of PET-guided SPECT in a bicentric study.","authors":"Solène Perret, Léa Albe, Agathe Edet-Sanson, David Tonnelet, Thomas Carlier, Camilla Kohi, Matthieu Barbaud, Romain Modzelewski, Pierre Vera, Laetitia Augusto, Frédéric Di Fiore, Sébastien Hapdey, Clément Bailly, Arnaud Dieudonné, Pierre Decazes","doi":"10.1186/s13550-026-01468-y","DOIUrl":null,"url":null,"abstract":"Background: Manual segmentation of prostate cancer metastases on PSMA PET/CT and SPECT/CT is time-consuming and poorly scalable, particularly in highly metastatic patients. This study evaluated nnU-Net-based automatic segmentation models trained on PET and SPECT either separately or jointly and assessed whether PET-derived information can improve SPECT lesion segmentation. Seventy-three patients with metastatic castration-resistant prostate cancer treated with ¹⁷⁷Lu-PSMA were retrospectively included: 48 from the Henri Becquerel Cancer Center (HBCC) and 25 from Nantes University Hospital (NUH). For each patient, ⁶⁸Ga-PSMA PET/CT and ¹⁷⁷Lu-PSMA SPECT/CT were acquired before and during the first treatment cycle respectively. All images were manually segmented by four nuclear medicine physicians in consensus. Four nnU-Net models were trained: M1 (PET/CT only), M2 (SPECT/CT only), M3 (joint PET/CT+SPECT/CT, unimodal input at inference), and M4 (SPECT/CT with PET/CT segmentation as a priori input). Models were first trained and internally validated on HBCC data, then retrained on the full HBCC cohort and externally validated on NUH data.Results: For PET/CT segmentation, M1 and M3 achieved comparable performance. M1 reached DSCs of 0.83 ± 0.19 (internal) and 0.76 ± 0.22 (external), while M3 achieved 0.83 ± 0.16 (internal) and 0.77 ± 0.21 (external). For SPECT/CT, the PET-guided model M4 (DSC: 0.63 ± 0.24 internal; 0.78 ± 0.14 external; PPV: 0.65 ± 0.26 internal; 0.75 ± 0.23 external) provided the best results. Compared with the SPECT-only model M2 (DSC: 0.61 ± 0.26 internal; 0.70 ± 0.25 external; PPV: 0.63 ± 0.25 internal; 0.71 ± 0.24 external), M4 showed no statistically significant difference in internal validation (DSC p = 0.35), while being statistically significant in external validation (DSC p = 0.014).Conclusion: The nnU-Net framework enables accurate lesion segmentation on both ⁶⁸Ga-PSMA PET/CT and ¹⁷⁷Lu-PSMA SPECT/CT. While PET-only and joint PET+SPECT models perform similarly on PET images, incorporating PET-derived segmentations as prior information tends to improve SPECT/CT lesion segmentation. This PET-guided SPECT segmentation strategy leverages the higher spatial resolution of PET and represents a key step towards fully automated extraction of volumetric and dosimetric biomarkers for personalized prostate cancer treatment.","PeriodicalId":11611,"journal":{"name":"EJNMMI Research","volume":" ","pages":""},"PeriodicalIF":2.7000,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13593981/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"EJNMMI Research","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s13550-026-01468-y","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Manual segmentation of prostate cancer metastases on PSMA PET/CT and SPECT/CT is time-consuming and poorly scalable, particularly in highly metastatic patients. This study evaluated nnU-Net-based automatic segmentation models trained on PET and SPECT either separately or jointly and assessed whether PET-derived information can improve SPECT lesion segmentation. Seventy-three patients with metastatic castration-resistant prostate cancer treated with ¹⁷⁷Lu-PSMA were retrospectively included: 48 from the Henri Becquerel Cancer Center (HBCC) and 25 from Nantes University Hospital (NUH). For each patient, ⁶⁸Ga-PSMA PET/CT and ¹⁷⁷Lu-PSMA SPECT/CT were acquired before and during the first treatment cycle respectively. All images were manually segmented by four nuclear medicine physicians in consensus. Four nnU-Net models were trained: M1 (PET/CT only), M2 (SPECT/CT only), M3 (joint PET/CT+SPECT/CT, unimodal input at inference), and M4 (SPECT/CT with PET/CT segmentation as a priori input). Models were first trained and internally validated on HBCC data, then retrained on the full HBCC cohort and externally validated on NUH data.

Results: For PET/CT segmentation, M1 and M3 achieved comparable performance. M1 reached DSCs of 0.83 ± 0.19 (internal) and 0.76 ± 0.22 (external), while M3 achieved 0.83 ± 0.16 (internal) and 0.77 ± 0.21 (external). For SPECT/CT, the PET-guided model M4 (DSC: 0.63 ± 0.24 internal; 0.78 ± 0.14 external; PPV: 0.65 ± 0.26 internal; 0.75 ± 0.23 external) provided the best results. Compared with the SPECT-only model M2 (DSC: 0.61 ± 0.26 internal; 0.70 ± 0.25 external; PPV: 0.63 ± 0.25 internal; 0.71 ± 0.24 external), M4 showed no statistically significant difference in internal validation (DSC p = 0.35), while being statistically significant in external validation (DSC p = 0.014).

Conclusion: The nnU-Net framework enables accurate lesion segmentation on both ⁶⁸Ga-PSMA PET/CT and ¹⁷⁷Lu-PSMA SPECT/CT. While PET-only and joint PET+SPECT models perform similarly on PET images, incorporating PET-derived segmentations as prior information tends to improve SPECT/CT lesion segmentation. This PET-guided SPECT segmentation strategy leverages the higher spatial resolution of PET and represents a key step towards fully automated extraction of volumetric and dosimetric biomarkers for personalized prostate cancer treatment.

查看原文本刊更多论文

6⁸Ga-PSMA PET/CT和¹⁷Lu-PSMA SPECT/CT的自动病灶分割：PET引导SPECT在双中心研究中的附加价值。

背景：在PSMA PET/CT和SPECT/CT上对前列腺癌转移进行人工分割耗时且难以扩展，特别是在高转移患者中。本研究评估了基于nnu - net的PET和SPECT单独或联合训练的自动分割模型，并评估了PET衍生信息是否可以改善SPECT病变分割。采用1⁷⁷Lu-PSMA治疗的73例转移性去雄抵抗性前列腺癌患者被回顾性纳入研究：48例来自Henri Becquerel癌症中心（HBCC）， 25例来自南特大学医院（NUH）。对于每位患者，在第一个治疗周期之前和期间分别获得了⁶⁸Ga-PSMA PET/CT和¹⁷Lu-PSMA SPECT/CT。所有图像均由四位核医学医师一致手动分割。训练了四个nnU-Net模型：M1（仅PET/CT）， M2（仅SPECT/CT）， M3 （PET/CT联合+SPECT/CT，推理时单峰输入）和M4 （SPECT/CT与PET/CT分割作为先验输入）。模型首先在HBCC数据上进行训练和内部验证，然后在整个HBCC队列上进行再训练，并在NUH数据上进行外部验证。结果：对于PET/CT分割，M1和M3的性能相当。M1达到DSCs为0.83±0.19（内）和0.76±0.22（外），M3达到0.83±0.16（内）和0.77±0.21（外）。对于SPECT/CT， pet引导的M4模型（DSC: 0.63±0.24内；0.78±0.14外；PPV: 0.65±0.26内；0.75±0.23外）效果最好。M4与仅spect模型M2 （DSC: 0.61±0.26内部；0.70±0.25外部；PPV: 0.63±0.25内部；0.71±0.24外部）相比，内部验证差异无统计学意义（DSC p = 0.35），外部验证差异有统计学意义（DSC p = 0.014）。结论：nnU-Net框架能够在26⁸Ga-PSMA PET/CT和¹⁷Lu-PSMA SPECT/CT上实现准确的病灶分割。虽然PET-only和PET+SPECT联合模型在PET图像上的表现相似，但将PET衍生的分割作为先验信息倾向于改善SPECT/CT病变分割。这种PET引导的SPECT分割策略利用了PET更高的空间分辨率，代表了个性化前列腺癌治疗中全自动提取体积和剂量生物标志物的关键一步。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

EJNMMI Research RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING&nb-

CiteScore

5.90

自引率

3.10%

发文量

审稿时长

13 weeks

期刊介绍： EJNMMI Research publishes new basic, translational and clinical research in the field of nuclear medicine and molecular imaging. Regular features include original research articles, rapid communication of preliminary data on innovative research, interesting case reports, editorials, and letters to the editor. Educational articles on basic sciences, fundamental aspects and controversy related to pre-clinical and clinical research or ethical aspects of research are also welcome. Timely reviews provide updates on current applications, issues in imaging research and translational aspects of nuclear medicine and molecular imaging technologies. The main emphasis is placed on the development of targeted imaging with radiopharmaceuticals within the broader context of molecular probes to enhance understanding and characterisation of the complex biological processes underlying disease and to develop, test and guide new treatment modalities, including radionuclide therapy.