基于[¹⁸F]FDG PET/CT的深度学习自动乳腺癌代谢分期与核医学医师及临床分期比较
本研究回顾性纳入403例经组织学确诊的乳腺癌患者,开发基于深度学习的[¹⁸F]FDG PET/CT自动代谢分期框架,并与核医学医师代谢分期及临床分期比较。结果显示DL分期与NM/临床分期在T、N、M的符合率分别为75/62%、87/74%、83/82%,在N3和M1中敏感性与特异性分别达0.86/0.96和0.97/0.78。结论认为该DL框架可作为乳腺癌代谢分期的辅助工具,但仍需专家监督。
要点
- DL自动分期与核医学医师代谢分期在T、N、M的符合率分别为75%、87%、83%。
- 对N3(extra-ALN)和M1(dM),DL基于NM参考的敏感性/特异性分别为0.86/0.96和0.97/0.78。
- NM代谢分期与临床分期在T、N、M的符合率分别为63%、80%、99%。
核心结论
- DL框架在乳腺癌[¹⁸F]FDG PET/CT代谢分期中展现作为辅助工具的潜力,但专家监督仍不可或缺。
研究设计与样本
- 回顾性纳入403例组织学确诊的乳腺癌患者,均接受全身[¹⁸F]FDG PET/CT分期。
- 数据按分层方法分为训练集303例和测试集100例,开发类别特异性DL分割网络并经后处理流程推导代谢TNM分期。
- 以核医学医师基于分割的代谢分期和临床分期为参考,计算T、N、M的准确性、敏感性和特异性,并用Dice系数和病灶检测指标评估分割性能。
主要发现
- DL分期与NM/临床分期的符合率在T为75%/62%,N为87%/74%,M为83%/82%。
- 对N3(存在extra-ALN)和M1(存在dM),DL基于NM参考的敏感性/特异性分别为0.86/0.96和0.97/0.78。
- 分割性能对pT、ALN和dM为良好至优秀(中位DC≥0.83,LD≥0.78),对extra-ALN为中等(中位DC=0.63,LD≥0.71)。
- NM代谢分期与临床分期在T、N、M的符合率分别为63%、80%、99%。
关键数值与推荐
- DL分期与NM/临床分期符合率:T 75%/62%,N 87%/74%,M 83%/82%。
- N3的DL敏感性/特异性(NM参考)为0.86/0.96,M1为0.97/0.78。
- 分割中位DC:pT、ALN、dM≥0.83,extra-ALN=0.63;中位LD:pT、ALN、dM≥0.78,extra-ALN≥0.71。
- NM代谢分期与临床分期符合率:T 63%,N 80%,M 99%。
临床怎么用
- 该DL框架可作为乳腺癌[¹⁸F]FDG PET/CT代谢分期的辅助工具,提升工作流效率。
- 在N3和M1等关键分期中DL表现出较高的敏感性和特异性,但最终分期仍需核医学医师专家监督。
局限性
- 原文未报告局限性相关内容。
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
PURPOSE: This study aimed to evaluate a deep-learning (DL)-based framework to automatically perform breast cancer (BC) metabolic staging on [¹⁸F]FDG PET/CT, and to assess agreement among DL-based, nuclear medicine (NM) physician-based, and clinical staging. METHODS: A total of 403 histologically confirmed BC patients who underwent whole-body staging [¹⁸F]FDG PET/CT were retrospectively included. All [18F]FDG avid lesions suspected of malignancy were segmented by an NM physician and classified into four key tumor regions: primary tumor (pT), regional axillary lymph nodes (ALN), extra-axillary locoregional nodes (extra-ALN), and distant metastases (dM). Data were split into training (n = 303) and testing (n = 100) sets using a stratified approach. Class-specific DL segmentation networks were developed. A post-processing pipeline was implemented to derive metabolic TNM staging from the DL-based segmentation. Using NM-based and clinical staging segmentation as reference, accuracy, sensitivity, and specificity were computed for T, N, and M. Segmentation performance was evaluated using the Dice coefficient (DC) and lesion detection (LD) metrics. NM-based metabolic staging was also compared with clinical staging. RESULTS: DL-based staging showed concordance with NM-based/clinical staging of 75/62%, 87/74%, and 83/82% for T, N, and M, respectively. For N3 (presence of extra-ALN) and M1 (presence of dM), where [¹⁸F]FDG PET/CT is particularly relevant, sensitivity/specificity of DL-based (NM-based reference) were 0.86/0.96, and 0.97/0.78, respectively. Segmentation performance was good to excellent for pT, ALN, and dM (median DC ≥ 0.83 and LD ≥ 0.78), and moderate for extra-ALN (median DC = 0.63 and LD ≥ 0.71). NM-based metabolic staging agreed with clinical staging in 63%, 80%, and 99% of cases for T, N, and M, respectively. CONCLUSION: Although expert supervision remains essential, the developed DL-based framework demonstrates potential as a supportive tool for metabolic staging in BC patients, facilitating a workflow-efficient [¹⁸F]FDG PET/CT-based staging assessment.
引用
- Constantino CS, Oliveira C, Oliveira FPM, et al. Automatic metabolic breast cancer staging using [¹⁸F]FDG PET/CT: comparison with nuclear medicine physician-based and clinical staging. Eur J Nucl Med Mol Imaging. 2026. PMID: 42319488. doi:10.1007/s00259-026-08005-y
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。