Page 64 - 南京医科大学自然版
P. 64

第46卷第4期
               ·532 ·                            南 京    医 科 大 学 学         报                        2026年4月


             [17]ZUBAIR M,OWAIS M,HASSAN T,et al. An interpreta⁃     framework for contrastive learning of visual representa⁃
                   ble framework for gastric cancer classification using multi⁃  tions[C]. Proceedings of the 37th International Confer⁃
                   channel attention mechanisms and transfer learning ap⁃  ence on Machine Learning,2020,119:1597-1607
                   proach on histopathology images[J]. Sci Rep,2025,15  [28]NAIMI A I,BALZER L B. Stacked generalization:an in⁃
                  (1):13087                                          troduction to super learning[J]. Eur J Epidemiol,2018,
             [18]LI L L,GENG Y,CHEN T,et al. Deep learning model tar⁃  33(5):459-464
                   geting cancer surrounding tissues for accurate cancer  [29]AJANI J A,D’AMICO T A,BENTREM D J,et al. Gastric
                   diagnosis based on histopathological images[J]. J Transl  cancer,version 2.2022,NCCN clinical practice guide⁃
                   Med,2025,23(1):110                                lines in oncology[J]. J Natl Compr Canc Netw,2022,20
             [19]ZHAO Y S,LI L S,YU X,et al. SurvGraph:a hybrid ⁃   (2):167-192
                   graph attention network for survival prediction using  [30]TOMITA N,ABDOLLAHI B,WEI J,et al. Attention ⁃
                   whole slide pathological images in gastric cancer[J]. Neu⁃  based deep neural networks for detection of cancerous
                   ral Netw,2025,189:107607                          and precancerous esophagus tissue on histopathological
             [20]TOGAÇAR M,ÖZKURT K B,ERGEN B,et al. Breast⁃         slides[J]. JAMA Netw Open,2019,2(11):e1914645
                   Net:a novel convolutional neural network model through  [31]SELVARAJU R R,COGSWELL M,DAS A,et al. Grad⁃
                   histopathological images for the diagnosis of breast cancer  CAM:visual explanations from deep networks via gradient⁃
                  [J]. Phys A Stat Mech Appl,2020,545:123592         based localization[J]. Int J Comput Vis,2020,128(2):
             [21]金怀平,薛飞跃,李振辉,等. 基于病理图像集成深度学                          336-359
                   习的胃癌预后预测方法[J]. 电子与信息学报,2023,45               [32]SANDMANN S,HEGSELMANN S,FUJARSKI M,et al.
                  (7):2623-2633                                      Benchmark evaluation of DeepSeek large language mod⁃
                   JIN H P,XUE F Y,LI Z H,et al. Gastric cancer progno⁃  els in clinical decision ⁃ making[J]. Nat Med,2025,31
                   sis prediction method based on integrated deep learning  (8):2546-2549
                   of pathological images[J]. Journal of Electronics & Infor⁃  [33]CRISTESCU R,LEE J,NEBOZHYN M,et al. Molecular
                   mation Technology,2023,45(7):2623-2633            analysis of gastric cancer identifies subtypes associated
             [22]陈园琼,邹北骥,张美华,等. 医学影像处理的深度学习                          with distinct clinical outcomes[J]. Nat Med,2015,21
                   可解释性研究进展[J]. 浙江大学学报(理学版),2021,                   (5):449-456
                   48(1):18-29,40                               [34]GAO F,DING J X,GAI B W,et al. Interpretable multi⁃
                   CHEN Y Q,ZOU B J,ZHANG M H,et al. Research prog⁃  modal fusion model for bridged histology and genomics
                   ress on deep learning interpretability in medical image  survival prediction in pan⁃cancer[J]. Adv Sci,2025,12
                   processing[J]. Journal of Zhejiang University(Science  (17):2407060
                   Edition),2021,48(1):18-29,40                 [35]ZOLKOV E,GINZBERG N,COHEN E. A mm ⁃ wave
             [23]CHOLLET F. Xception:deep learning with depthwise sep⁃  blocker ⁃ tolerant receiver achieving <4 dB NF and - 3.5
                   arable convolutions[C]//2017 IEEE Conference on Com⁃  dBm B1dB in 65⁃nm CMOS[C]. 2023 IEEE Radio Fre⁃
                   puter Vision and Pattern Recognition(CVPR). Honolulu,  quency Integrated Circuits Symposium(RFIC),2023:
                   HI,USA. IEEE,2017:1800-1807                       297-300
             [24]SANDLER M,HOWARD A,ZHU M L,et al. Mobile⁃      [36]LEE K,CAVALCANTI T C,KIM S,et al. Multi⁃task and
                   NetV2:inverted residuals and linear bottlenecks[C]//  few ⁃ shot learning ⁃ based fully automatic deep learning
                   2018 IEEE/CVF Conference on Computer Vision and Pat⁃  platform for mobile diagnosis of skin diseases[J]. IEEE J
                   tern Recognition. Salt Lake City,UT,USA. IEEE,2018:  Biomed Health Inform,2023,27(1):176-187
                   4510-4520                                    [37]张建新,高程阳,孙        鉴,等. 混合监督学习的乳腺癌全
             [25]RAGHU M,ZHANG C Y,KLEINBERG J M,et al. Trans⁃       切片病理图像分类[J]. 中国图象图形学报,2024,29
                   fusion:understanding transfer learning for medical imag⁃  (9):2825-2836
                   ing[C].//Neural Information Processing Systems.,2019  ZHANG J X,GAO C Y,SUN J,et al. Breast cancer whole
             [26]TELLEZ D,LITJENS G,VAN DER LAAK J,et al. Neu⁃       slide pathological image classification based on hybrid
                   ral image compression for gigapixel histopathology image  supervised learning[J]. Journal of Image and Graphics,
                   analysis[J]. IEEE Trans Pattern Anal Mach Intell,2021,  2024,29(9):2825-2836
                   43(2):567-578                                 (收稿:2025-03-26;修回:2025-09-17;录用:2025-09-19)
             [27]CHEN T,KORNBLITH S,NOROUZI M,et al. A simple                                 (本文编辑:戴王娟)
   59   60   61   62   63   64   65   66   67   68   69