切换至 "中华医学电子期刊资源库"

中华结直肠疾病电子杂志 ›› 2025, Vol. 14 ›› Issue (05) : 445 -456. doi: 10.3877/cma.j.issn.2095-3224.2025.05.009

论著

基于磁共振的深度学习重建方法在直肠癌术前评估中的应用研究
鲁莽1, 马晓璐1, 沈浮1, 王颢2, 邵成伟1, 张卫2, 陆建平1, 陆海迪1,()   
  1. 1200433 海军军医大学第一附属医院(上海长海医院)影像医学科
    2200433 海军军医大学第一附属医院(上海长海医院)肛肠外科
  • 收稿日期:2025-02-13 出版日期:2025-10-25
  • 通信作者: 陆海迪

MRI-based deep learning reconstruction in preoperative evaluating of rectal cancer

Mang Lu1, Xiaolu Ma1, Fu Shen1, Hao Wang2, Chengwei Shao1, Wei Zhang2, Jianping Lu1, Haidi Lu1,()   

  1. 1Department of Radiology, the First Affiliated Hospital of Naval Medical University (Changhai Hospital), Shanghai 200433, China
    2Department of Colorectal Surgery, the First Affiliated Hospital of Naval Medical University (Changhai Hospital), Shanghai 200433, China
  • Received:2025-02-13 Published:2025-10-25
  • Corresponding author: Haidi Lu
引用本文:

鲁莽, 马晓璐, 沈浮, 王颢, 邵成伟, 张卫, 陆建平, 陆海迪. 基于磁共振的深度学习重建方法在直肠癌术前评估中的应用研究[J/OL]. 中华结直肠疾病电子杂志, 2025, 14(05): 445-456.

Mang Lu, Xiaolu Ma, Fu Shen, Hao Wang, Chengwei Shao, Wei Zhang, Jianping Lu, Haidi Lu. MRI-based deep learning reconstruction in preoperative evaluating of rectal cancer[J/OL]. Chinese Journal of Colorectal Diseases(Electronic Edition), 2025, 14(05): 445-456.

目的

探讨深度学习重建(DLR)技术对图像质量以及直肠癌术前诊断效能的影响。

方法

回顾性分析2017年7月至2022年3月海军军医大学第一附属医院(上海长海医院)术后病理证实为直肠癌的患者临床、病理及治疗前磁共振成像(MRI)资料。采用DLR对MRI中的T2WI、DWI和CE-T1WI序列进行重建优化,并由5名不同诊断经验医师分别对原始MRI和DLR-MRI图像集进行图像质量评估。通过计算病变的信噪比(SNR)和对比噪声比(CNR)对图像进行客观定量评估,并对整体图像质量和病变显示性能进行图像主观定性分析。对两组图像的术前TN分期进行主观诊断以及深度学习模型诊断效能比较。

结果

最终纳入178例患者。DLR-MRI图像三个序列在客观定量指标(SNR和CNR)和主观诊断(整体图像质量和病变显示表现)方面均显著高于原始图像(P<0.0001)。在MRI图像诊断效能方面,DLR-MRI提高了医师3、5对直肠癌术前T分期曲线下面面积(AUC)值,最高达0.921(医师3),但差异无统计学意义(P>0.05),其中医师4的DLR-MRI诊断效能提升显著,差异具有统计学意义(Z=2.971,P=0.003),而且DLR-MRI也显著提升了深度学习分类模型的诊断效能,AUC值由0.797提升至0.937(Z=2.505,P=0.043)。然而,无论主观诊断还是深度学习模型诊断,在直肠癌术前N分期方面的改善差异均无统计学意义(P>0.05)。进一步DCA显示,在直肠癌术前T分期中,基于DLR技术的深度学习模型展现出更高的临床净收益,而在直肠癌术前N分期方面,两个模型临床效益较为相似。

结论

DLR技术可以优化MRI图像质量以增强病变显示能力,进而基于DLR的深度学习分类模型可以一定程度提升直肠癌术前诊断效能。

Objective

To explore the impact of deep learning reconstruction (DLR) technology on image quality and the diagnostic performance of preoperative rectal cancer.

Methods

A retrospective analysis was conducted on patients with rectal cancer confirmed by postoperative pathology in our hospital from July 2017 to March 2022, focusing on their clinical, pathological, and pre-treatment MRI data. The DLR technology was used to reconstruct and optimize the T2WI, DWI, and CE-T1WI sequences in MRI. Five readers with different diagnostic experiences separately assessed the image quality of the original-MRI and DLR-MRI image sets. Objective quantitative evaluation of the images was conducted by calculating the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the lesions, while subjective qualitative analysis was performed on overall image quality and lesion display performance. Finally, subjective diagnosis of preoperative TN staging and comparison of diagnostic performance of deep learning models were performed for the two groups of images.

Results

A total of 178 patients were included. The DLR-MRI images showed significantly higher values in both objective quantitative metrics (SNR and CNR) and subjective evaluations (overall image quality and lesion display performance) compared to the original images (P<0.0001). In terms of the diagnostic efficacy of MRI images, DLR-MRI increased the AUC values of preoperative T staging for rectal cancer in readers 3 and 5, reaching up to 0.921 (reader 3), but the difference was not statistically significant (P>0.05). Among them, the diagnostic efficacy of reader 4 was significantly improved (Z=2.971, P=0.003). Moreover, DLR-MRI significantly enhanced the diagnostic efficacy of the deep learning classification model, with the AUC value increasing from 0.797 to 0.937 (Z=2.505, P=0.043). However, neither the subjective assessment nor the DL classification model achieved statistically significant improvements in preoperative N-staging for rectal cancer. Further analysis using DCA showed that in preoperative T-staging of rectal cancer, the DLR-DL model demonstrated higher clinical net benefit, while the clinical benefits of the two models were similar in preoperative staging.

Conclusion

DLR technology can optimize MRI image quality to enhance lesion display capabilities. Furthermore, the deep learning classification model based on DLR can improve the preoperative diagnostic performance of rectal cancer to a certain extent.

表1 直肠MRI扫描的主要参数
图1 原始图像和DLR图像比较(女性,61岁,直肠腺癌,pT3N2)。1A和1D分别为原始T2WI和DLR-T2WI:1B和1E分别为原始DWI和DLR-DWI;1C和1F分别为原始CE-T1WI和DLR-CE-T1WI
表2 患者一般资料[±s,例(%)]
表3 原始MRI和DLR-MRI图像集的SNR和CNR的比较
图2 医师1、2及其均值的原始MRI及DLR-MRI图像SNR(2A)和CNR(2B)比较图
表4 原始MRI和DLR-MRI图像集主观图像质量得分及其比较
图3 三位医师对原始MRI和DLR-MRI图像主观图像质量得分的堆叠条形图。3A:为T2WI序列;3B:为DWI序列;3C:为CE-T1WI序列。注:DLR代表深度学习重建图像集;Original代表原始图像集。Reader 3、4和5分别代表工作经验为10年、5年和1年的医师
表5 原始和DLR MRI图像集评估直肠癌术前T分期的主观诊断效能
表6 原始和DLR MRI图像集评估直肠癌术前N分期的主观诊断效能
图4 医师主观评价的ROC曲线图。4A为术前MR-T分期;4B为术前MR-N分期。注:DLR代表深度学习重建图像集;Original代表原始图像集。Reader 3、4和5分别代表工作经验为10年、5年和1年的医师
表7 测试集原始和DLR-MRI图像集评估直肠癌术前TN分期的深度学习模型诊断效能
图5 深度学习诊断模型在测试集中ROC曲线图。5A为深度学习分类模型术前T分期;5B为深度学习分类模型术前N分期
图6 两个模型在测试集中的DCA图。蓝色和红色的曲线分别代表DLR-深度学习模型和原始-深度学习模型。6A为深度学习分类模型判断T分期的DCA图;6B为深度学习分类模型判断N分期的DCA图
图7 DL分类模型测试集中可视化分析展示图。7A为直肠癌N1-2期及N0期原图、融合图及类激活图展示,其中,直肠癌N1-2期预测类激活图显影明显,且结构分层明显,关注于病灶的核心区域;而阴性图像的类激活图结构分层不清,且显影相对较浅。7B为直肠癌T1-2期及T3-4期原图、融合图及类激活图展示,直肠癌T3-4期预测类激活图显影明显,且结构分层明显;而阴性图像的类激活图则结构为显示
[1]
Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024[J]. CA Cancer J Clin, 2024, 74(1): 12-49.
[2]
中国医师协会结直肠肿瘤专业委员会. 结直肠癌腹膜转移诊治专家共识(2025版)[J]. 中华胃肠外科杂志, 2025, 28(5): 441-449.
[3]
Jayaprakasam VS, Alvarez J, Omer DM, et al. Watch-and-wait approach to rectal cancer: the role of imaging[J]. Radiology, 2023, 307(1): e221529.
[4]
杨鋆,辛城霖,张忠涛. 中低位直肠癌的精准诊断与规范治疗[J]. 中华消化外科杂志, 2024, 23(1): 85-90.
[5]
Mi M, Weng S, Xu Z, et al. CSCO guidelines for colorectal cancer version 2023: updates and insights[J]. Chin J Cancer Res, 2023, 35(3): 233-238.
[6]
Benson AB, Venook AP, Al-Hawary MM, et al. Rectal Cancer, Version 2.2022, NCCN clinical practice guidelines in oncology[J]. J Natl Compr Canc Netw, 2022, 20(10): 1139-1167.
[7]
Peng W, Wan L, Tong X, et al. Prospective and multi-reader evaluation of deep learning reconstruction-based accelerated rectal MRI: image quality, diagnostic performance, and reading time[J]. Eur Radiol, 2024, 34(11): 7438-7449.
[8]
严福华. 深度学习MRI重建算法的临床应用和发展前景[J]. 磁共振成像, 2023, 14(5): 8-10.
[9]
Lebel RM. Performance characterization of a novel deep learning-based MR image reconstruction pipeline[EB/OL]. [2022-11-07](2025-02-13).

URL    
[10]
Kim M, Park T, Oh BY, et al. Performance reporting design in artificial intelligence studies using image-based TNM staging and prognostic parameters in rectal cancer: a systematic review[J]. Ann Coloproctol, 2024, 40(1): 13-26.
[11]
Oscanoa JA, Middione MJ, Alkan C, et al. Deep learning-based reconstruction for cardiac MRI: a review[J]. Bioengineering (Basel), 2023, 10(3): 334.
[12]
Johnson PM, Lin DJ, Zbontar J, et al. Deep learning reconstruction enables prospectively accelerated clinical knee MRI[J]. Radiology, 2023, 307(2): e220425.
[13]
Tsuboyama T, Onishi H, Nakamoto A, et al. Impact of deep learning reconstruction combined with a sharpening filter on single-shot fast spin-echo T2-weighted magnetic resonance imaging of the uterus[J]. Invest Radiol, 2022, 57(6): 379-386.
[14]
Park JC, Park KJ, Park MY, et al. Fast T2-weighted imaging with deep learning-based reconstruction: evaluation of image quality and diagnostic performance in patients undergoing radical prostatectomy[J]. J Magn Reson Imaging, 2022, 55(6): 1735-1744.
[15]
Cui J, Liu S, Tian Z, et al. ResLT: residual learning for long-tailed recognition[J]. IEEE Trans Pattern Anal Mach Intell, 2023, 45(3): 3695-3706.
[16]
Lee S, Palt S, Ma J, et al. Rectal MR imaging[J]. Radiol Clin North Am, 2025, 63(3): 419-434.
[17]
Nougaret S, Reinhold C, Mikhael HW, et al. The use of MR imaging in treatment planning for patients with rectal carcinoma: have you checked the "DISTANCE" ?[J]. Radiology, 2013, 268(2): 330-344.
[18]
Wetzel A, Viswanath S, Gorgun E, et al. Staging and restaging of rectal cancer with MRI: a pictorial review[J]. Semin Ultrasound CT MR, 2022, 43(6): 441-454.
[19]
Tomita H, Deguchi Y, Fukuchi H, et al. Combination of compressed sensing and parallel imaging for T2-weighted imaging of the oral cavity in healthy volunteers: comparison with parallel imaging[J]. Eur Radiol, 2021, 31(8): 6305-6311.
[20]
Cristobal-Huerta A, Poot D, Vogel MW, et al. Compressed sensing 3D-grase for faster high-resolution MRI[J]. Magn Reson Med, 2019, 82(3): 984-999.
[21]
Ishimoto Y, Ide S, Watanabe K, et al. Usefulness of pituitary high-resolution 3D MRI with deep-learning-based reconstruction for perioperative evaluation of pituitary adenomas[J]. Neuroradiology, 2024, 66(6): 937-945.
[22]
Wang S, Cao G, Wang Y, et al. Review and prospect: artificial intelligence in advanced medical imaging[J]. Front Radiol, 2021, 1: 781868.
[23]
Tang H, Hong M, Yu L, et al. Deep learning reconstruction for lumbar spine MRI acceleration: a prospective study[J]. Eur Radiol Exp, 2024, 8(1): 67.
[24]
Zhu L, Shi B, Ding B, et al. Accelerated T2W imaging with deep learning reconstruction in staging rectal cancer: a preliminary study[J]. J Imaging Inform Med, 2024, 12: 11.
[25]
Bae SH, Hwang J, Hong SS, et al. Clinical feasibility of accelerated diffusion weighted imaging of the abdomen with deep learning reconstruction: comparison with conventional diffusion weighted imaging[J]. Eur J Radiol, 2022, 154: 110428.
[26]
Tanabe M, Higashi M, Yonezawa T, et al. Feasibility of high-resolution magnetic resonance imaging of the liver using deep learning reconstruction based on the deep learning denoising technique[J]. Magn Reson Imaging, 2021, 80: 121-126.
[27]
Kim B, Lee CM, Jang JK, et al. Deep learning-based imaging reconstruction for MRI after neoadjuvant chemoradiotherapy for rectal cancer: effects on image quality and assessment of treatment response[J]. Abdom Radiol (NY), 2023, 48(1): 201-210.
[28]
Gudbjartsson H, Patz S. The rician distribution of noisy MRI data[J]. Magn Reson Med, 1995, 34(6): 910-914.
[29]
Hou M, Zhou L, Sun J. Deep-learning-based 3D super-resolution MRI radiomics model: superior predictive performance in preoperative T-staging of rectal cancer[J]. Eur Radiol, 2023, 33(1): 1-10.
[30]
Borgheresi A, De Muzio F, Agostini A, et al. Lymph nodes evaluation in rectal cancer: where do we stand and future perspective[J]. J Clin Med, 2022,11(9): 2599
[31]
Stijns R, Philips B, Nagtegaal ID, et al. USPIO-enhanced MRI of lymph nodes in rectal cancer: a node-to-node comparison with histopathology[J]. Eur J Radiol, 2021, 138: 109636.
[32]
Langman G, Patel A, Bowley DM. Size and distribution of lymph nodes in rectal cancer resection specimens[J]. Dis Colon Rectum, 2015, 4(50): 406-414.
[33]
Zhu H, Zhang X, Li X, et al. Prediction of pathological nodal stage of locally advanced rectal cancer by collective features of multiple lymph nodes in magnetic resonance images before and after neoadjuvant chemoradiotherapy[J]. Chin J Cancer Res, 2019, 31(6): 984-992.
[34]
Feng W, Zhu L, Xia Y, et al. Deep learning-based reconstruction: a reliability assessment in preoperative magnetic resonance imaging for primary rectal cancer[J]. Quant Imaging Med Surg, 2024, 14(12): 8927-8941.
[1] 江瑶, 蒋程, 余翔, 谭莹, 温昕, 温慧莹, 彭桂艳, 李胜利. 基于注意力机制改进的子宫解剖结构检测与分割多任务模型的性能评估[J/OL]. 中华医学超声杂志(电子版), 2025, 22(08): 703-710.
[2] 严征远, 张恒, 曹能琦, 方兴超, 陈大敏. 单孔+1腹腔镜结直肠癌根治切除术的有效性及安全性临床观察[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(06): 615-618.
[3] 杨春燕, 周晓苹. 机器人辅助技术在腹腔镜结直肠癌根治术中的研究进展[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(05): 584-588.
[4] 党立力, 张流, 周鹏, 周华友, 赵伟. 保留左结肠动脉(升支)腹腔镜直肠癌全系膜切除术的临床观察[J/OL]. 中华普外科手术学杂志(电子版), 2025, 19(04): 442-445.
[5] 王明媚, 李勇. 肾盂癌的影像诊断及进展[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2025, 19(04): 412-417.
[6] 郑哲宇, 张磊, 张大伟, 潘卫东, 黄晓明. 全腹腔镜下ALPPS治疗结直肠癌肝转移的安全性和疗效[J/OL]. 中华肝脏外科手术学电子杂志, 2025, 14(05): 748-753.
[7] 鲁旭, 李华. 结直肠癌肝转移肝移植治疗研究进展[J/OL]. 中华肝脏外科手术学电子杂志, 2025, 14(04): 508-514.
[8] 关旭, 杨明. 中国微创手术的光辉历程与经自然腔道取标本手术的革新之路[J/OL]. 中华结直肠疾病电子杂志, 2025, 14(05): 385-388.
[9] 中国医师协会外科医师分会, 中华医学会外科分会胃肠外科学组, 中华医学会外科分会结直肠外科学组, 中国抗癌协会大肠癌专业委员会, 中国医师协会结直肠肿瘤专业委员会, 中国临床肿瘤学会结直肠癌专家委员会, 中国医师协会外科医师分会结直肠外科医师委员会, 中国医师协会肛肠医师分会肿瘤转移委员会, 中华医学会肿瘤学分会结直肠肿瘤学组, 中国医疗保健国际交流促进会转移肿瘤治疗学分会, 中国医疗保健国际交流促进会结直肠病分会. 结直肠癌肝转移诊断和综合治疗指南(V 2025)[J/OL]. 中华结直肠疾病电子杂志, 2025, 14(05): 398-411.
[10] 中国抗癌协会NOSES专业委员会, 中国抗癌协会大肠癌专业委员会. 肿瘤整合诊治技术指南·NOSES技术(结直肠癌部分)[J/OL]. 中华结直肠疾病电子杂志, 2025, 14(05): 412-416.
[11] 赵南, 张明凯, Bhargava Divija, 赵世光, 张大明. 结直肠癌脑转移的临床特征与治疗策略进展[J/OL]. 中华结直肠疾病电子杂志, 2025, 14(05): 427-435.
[12] 赵发宽, 向岩帕保, 杨家成, 梁伟纲, 张金源. 单孔+1腹腔镜下左半结肠癌根治术一例(附视频)[J/OL]. 中华结直肠疾病电子杂志, 2025, 14(04): 380-384.
[13] 王柯云, 孙雅佳, 李甜, 张钰哲, 郑颖, 张伟光, 王倩, 董哲毅. 糖尿病肾脏疾病早期发生风险预测模型的研究进展[J/OL]. 中华肾病研究电子杂志, 2025, 14(04): 218-225.
[14] 鲁旭娟, 孔振兴. miR-146a靶向PTP1B对结直肠癌细胞增殖、迁移及侵袭的影响[J/OL]. 中华消化病与影像杂志(电子版), 2025, 15(04): 294-299.
[15] 王美, 赵勇, 张健, 张俐娜, 丁健华, 曹煜. 基于CT测量肾周脂肪面积对Lap-ISR吻合口并发症的预测价值[J/OL]. 中华临床医师杂志(电子版), 2025, 19(05): 367-373.
阅读次数
全文


摘要


AI


AI小编
你好!我是《中华医学电子期刊资源库》AI小编,有什么可以帮您的吗?