MRI深度学习量化预测进展期胃癌隐匿性腹膜转移

MRI-based deep learning models for quantitative prediction of occult peritoneal metastasis in advanced gastric cancer

摘要

目的：探讨基于MRI的多种深度学习（DL）模型预测进展期胃癌隐匿性腹膜转移（OPM）的可行性及价值。方法：回顾性纳入476例术前MRI诊断为腹膜转移阴性、经手术病理确诊的进展期胃癌患者，按7:3随机分为训练集（334例）和测试集（142例）。在T₂WI-FS序列上手动勾画肿瘤三维感兴趣区，训练ResNet50、Med-ViT和TP-Mamba 3种DL模型，采用受试者工作特征曲线下面积（AUC）、敏感度、特异度、DeLong检验、综合判别改进指数（IDI）及临床决策曲线（DCA）评估效能。结果：Med-ViT模型（训练集AUC=0.860；测试集AUC=0.763）预测OPM的效能均优于TP-Mamba（训练集AUC=0.767；测试集AUC=0.652）及ResNet50模型（训练集AUC=0.528；测试集AUC=0.570）。在训练集及测试集中，DeLong检验显示Med-ViT的AUC值均显著高于其余模型（P<0.05），IDI分析显示Med-ViT模型综合判别改善能力最优，DCA表明其净获益最高。结论：基于自注意力机制的Med-ViT深度学习模型可准确预测进展期胃癌OPM风险，效能显著优于ResNet50与TP-Mamba，有望成为术前无创筛查OPM高危患者的辅助工具。

Abstract

Objective To explore the feasibility and value of multiple deep learning models based on magnetic resonance imaging (MRI) for the prediction of occult peritoneal metastasis (OPM) in advanced gastric cancer. Methods A retrospective analysis was conducted on 476 patients with pathologically confirmed advanced gastric cancer and negative preoperative MRI findings for peritoneal metastasis. All patients were randomly divided into a training set (n=334) and a test set (n=142) at a ratio of 7:3. Three-dimensional tumor regions of interest were manually delineated on T₂WI-FS sequences. Three deep learning models, including ResNet50, Med-ViT, and TP-Mamba, were trained, and their performance was evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, the DeLong test, integrated discrimination improvement, and decision curve analysis. Results Med-ViT model (training set AUC=0.860; test set AUC=0.763) exhibited superior performance in predicting OPM compared with TP-Mamba model (training set AUC=0.767; test set AUC=0.652) and ResNet50 model (training set AUC=0.528; test set AUC=0.570). In both the training and test sets, the DeLong test showed that Med-ViT had significantly higher AUC values than the other two models (P<0.05). Integrated discrimination improvement analysis demonstrated that Med-ViT model achieved the optimal comprehensive discrimination improvement ability, and decision curve analysis reflected its greatest net benefit. Conclusion The self-attention mechanism-based Med-ViT deep learning model can accurately predict the risk of OPM in advanced gastric cancer and outperforms ResNet50 and TP-Mamba. Therefore, it is expected to serve as a reliable auxiliary tool for non-invasive preoperative screening of patients at high risk of OPM.

关键词

胃癌 / 隐匿性腹膜转移 / 磁共振成像 / 深度学习

Key words

gastric cancer / occult peritoneal metastasis / magnetic resonance imaging / deep learning

参考文献

[1] Chen WQ, Zheng RS, Baade PD, et al. Cancer statistics in China, 2015[J]. CA Cancer J Clin, 2016, 66(2): 115-132.

[2] Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249.

[3] Siegel RL, Giaquinto AN, Jemal A . Cancer statistics, 2024[J]. CA Cancer J Clin, 2024, 74(1): 12-49.

[4] Wang YA, Yan QJ, Fan CM, et al. Overview and countermeasures of cancer burden in China[J]. Sci China Life Sci, 2023, 66(11): 2515-2526.

[5] Cheng XD, Dai EY, Wu JB, et al. Atlas of metastatic gastric cancer links ferroptosis to disease progression and immunotherapy response[J]. Gastroenterology, 2024, 167(7): 1345-1357.

[6] Paul N, Surendran S, Yacob M, et al. Occult omental metastasis in gastric adenocarcinoma: an analysis of incidence, predictors, and outcomes[J]. South Asian J Cancer, 2022, 11(4): 299-308.

[7] Thomassen I, van Gestel YR, van Ramshorst B, et al. Peritoneal carcinomatosis of gastric origin: a population—based study on incidence, survival and risk factors[J]. Int J Cancer, 2014, 134(3): 622-628.

[8] Fujitani K, Yang HK, Mizusawa J, et al. Gastrectomy plus chemotherapy versus chemotherapy alone for advanced gastric cancer with a single non—curable factor (REGATTA): a phase 3, randomised controlled trial [J]. Lancet Oncol, 2016, 17(3): 309-318.

[9] Salati M, Valeri N, Spallanzani A, et al. Oligometastatic gastric cancer: an emerging clinical entity with distinct therapeutic implications[J]. Eur J Surg Oncol, 2019, 45(8): 1479-1482.

[10] Kim SJ, Kim HH, Kim YH, et al. Peritoneal metastasis: detection with 16— or 64—detector row CT in patients undergoing surgery for gastric cancer[J]. Radiology, 2009, 253(2): 407-415.

[11] Ng D, Cyr D, Khan S, et al. Molecular mechanisms of metastatic peritoneal dissemination in gastric adenocarcinoma[J]. Cancer Metastasis Rev, 2025, 44(2): 50.

[12] Lin L, Zhang P, Wang YY, et al. Early vs conventional initiation of adjuvant chemotherapy in advanced gastric cancer: a propensity—matched outcomes study[J]. World J Gastroenterol, 2025, 31(42): 110069.

[13] Burbidge S, Mahady K, Naik K . The role of CT and staging laparoscopy in the staging of gastric cancer[J]. Clin Radiol, 2013, 68(3): 251-255.

[14] Buia A, Stockhausen F, Hanisch E . Laparoscopic surgery: a qualified systematic review[J]. World J Methodol, 2015, 5(4): 238-254.

[15] Hu QC, Zhang SY, Yang K, et al. 68Ga—FAPI—04 PET for detecting occult peritoneal metastasis in locally advanced gastric cancer: diagnostic performance and cost analyses in a single—center, prospective cohort study[J]. J Nucl Med, 2026, 67(1): 53-59.

[16] Mao Q, Zhou MT, Zhao ZP, et al. Role of radiomics in the diagnosis and treatment of gastrointestinal cancer[J]. World J Gastroenterol, 2022, 28(42): 6002-6016.

[17] Dong D, Tang L, Li ZY, et al. Development and validation of an individualized nomogram to identify occult peritoneal metastasis in patients with advanced gastric cancer[J]. Ann Oncol, 2019, 30(3): 431-438.

[18] Huang J, Chen YD, Zhang YY, et al. Comparison of clinical—computed tomography model with 2D and 3D radiomics models to predict occult peritoneal metastases in advanced gastric cancer[J]. Abdom Radiol (NY), 2022, 47(1): 66-75.

[19] Li QW, Jiang ZJ, Zhu Y, et al. CT—based scores for extramural vascular invasion and occult peritoneal metastasis correlate with gastric cancer survival[J]. Eur Radiol, 2025, 35(9): 5733-5747.

[20] Wang LL, Lü P, Xue Z, et al. Novel CT based clinical nomogram comparable to radiomics model for identification of occult peritoneal metastasis in advanced gastric cancer[J]. Eur J Surg Oncol, 2022, 48(10): 2166-2173.

[21] Park CM, Lee JH . Deep learning for lung cancer nodal staging and real—world clinical practice[J]. Radiology, 2022, 302(1): 212-213.

[22] Zhu ZN, Feng QX, Li Q, et al. Machine learning—based CT radiomics approach for predicting occult peritoneal metastasis in advanced gastric cancer preoperatively[J]. Clin Radiol, 2025, 80: 106727.

[23] Zhou SK, Greenspan H, Davatzikos C, et al. A review of deep learning in medical imaging: imaging traits, technology trends, case studies with progress highlights, and future promises[J]. Proc IEEE, 2021, 109(5): 820-838.

[24] Jiang YM, Liang XK, Wang W, et al. Noninvasive prediction of occult peritoneal metastasis in gastric cancer using deep learning[J]. JAMA Netw Open, 2021, 4(1): e2032269.

[25] Li ML, Sun K, Dai WX, et al. Preoperative prediction of peritoneal metastasis in colorectal cancer using a clinical—radiomics model[J]. Eur J Radiol, 2020, 132: 109326.

[26] Wang XY, Wei MX, Chen Y, et al. Intratumoral and peritumoral MRI—based radiomics for predicting extrapelvic peritoneal metastasis in epithelial ovarian cancer[J]. Insights Imaging, 2024, 15(1): 281.

基金资助

广东省医学科研基金(B2023445)

梅州市社会发展科技计划项目(2023B19)