Page 66 - 《中国医疗器械杂志》2026年第2期

Chinese Journal of Medical Instrumentation 2026年 第50卷 第2期

综 合 评 述

[82] NIV Y, JOEL D, MEILIJSON I, et al. Evolution of arxiv.org/abs/2102.04306.
reinforcement learning in foraging bees: a simple [97] KAWAHARA J, BROWN C J, MILLER S P, et al.
explanation for risk averse behavior[J]. Neuro- BrainNetCNN: convolutional neural networks for brain
computing, 2002, 44-46: 951-956. networks; towards predicting neurodevelopment[J].
[83] SHIH C H, NAUGHTON N, HALDER U, et al. NeuroImage, 2017, 146: 1038-1049.
Hierarchical control and learning of a foraging [98] SAEEDINIA S A, JAHED-MOTLAGH M R,
CyberOctopus[J]. Adv Intell Syst, 2023, 5(9): 2300088. TAFAKHORI A, et al. Diagnostic biomarker discovery
[84] LITJENS G, KOOI T, BEJNORDI B E, et al. A survey from brain EEG data using LSTM, reservoir-SNN, and
on deep learning in medical image analysis[J]. Med NeuCube methods in a pilot study comparing epilepsy
Image Anal, 2017, 42: 60-88. and migraine[J]. Sci Rep, 2024, 14(1): 10667.
[85] ESTEVA A, KUPREL B, NOVOA R A, et al. [99] TUDOSIU P D, PINAYA W H L, FERREIRA
Dermatologist-level classification of skin cancer with DACOSTA P, et al. Realistic morphology-preserving
deep neural networks[J]. Nature, 2017, 542(7639): 115- generative modelling of the brain[J]. Nat Mach Intell,
118. 2024, 6(7): 811-819.
[86] BARATA C, ROTEMBERG V, CODELLA N C F, et [100] LI X, LI H, MA L. Continual learning of medical image
al. A reinforcement learning model for AI-based classification based on feature replay[C]//2022 16th
decision support in skin cancer[J]. Nat Med, 2023, IEEE International Conference on Signal Processing.
29(8): 1941-1946. Vol 1. IEEE, 2022: 426-430.
[87] MAO W, CHEN Y Z, HE Z B, et al. Brain structural [101] ZHENG G, ZHOU S, BRAVERMAN V, et al. Selective
connectivity guided vision transformers for experience replay compression using coresets for
identification of functional connectivity characteristics lifelong deep reinforcement learning in medical
in preterm neonates[J]. IEEE J Biomed Health Inform, imaging[C]//Proceedings of Machine Learning
2024, 28(4): 2223-2234. Research. Vol. 227. Cham: Springer Nature Switzerland,
[88] TIU E, TALIUS E, PATEL P, et al. Expert-level 2024: 1751-1764.
detection of pathologies from unannotated chest X-ray [102] YAMAZAKI K, VO-HO V K, BULSARA D, et al.
images via self-supervised learning[J]. Nat Biomed Eng, Spiking neural networks and their applications: a
2022, 6(12): 1399-1406. review[J]. Brain Sci, 2022, 12(7): 863.
[89] PARK S, KIM G, OH Y, et al. Self-evolving vision [103] CHEN S, CHEN Y, WANG Z, et al. A unified
transformer for chest X-ray diagnosis through geometric space bridging AI models and the human
knowledge distillation[J]. Nat Commun, 2022, 13(1):
brain[EB/OL]. arXiv preprint, 2025[2025-11-05].
3848.
https://arxiv.org/abs/2510.24342v1.
[90] ZHANG Y F, FENG W, WU Z Y, et al. Deep-learning
[104] CHEN Y Z, DU Y, XIAO Z X, et al. A unified and
model of ResNet combined with CBAM for
biologically plausible relational graph representation of
malignant–benign pulmonary nodules classification on
vision transformers[J]. IEEE Trans Neural Netw Learn
computed tomography images[J]. Medicina(Kannas),
Syst, 2025, 36(2): 3231-3243.
2023, 59(6): 1088.
[105] HUANG S C, JENSEN M, YEUNG-LEVY S, et al.
[91] DE FAUW J, LEDSAM J R, ROMERA-PAREDES B,
Multimodal foundation models for medical imaging –
et al. Clinically applicable deep learning for diagnosis
a systematic review and implementation guidelines
and referral in retinal disease[J]. Nat Med, 2018, 24(9):
[EB/OL]. medRxiv Preprint. (2024-10-23) [2025-08-04].
1342-1350.
https://doi.org/10.1101/2024.10.23.24316003.
[92] ALMAHASNEH M, XIE X, PAIEMENT A. AttentNet:
[106] XIAO Y, LIU Y Z, ZHANG B H, et al. Bio-plausible
fully convolutional 3D attention for lung nodule
reconfigurable spiking neuron for neuromorphic
detection[J]. SN Comput Sci, 2025, 6(3): 292.
computing[J]. Sci Adv, 2025, 11(6): eadr6733.
[93] MA C, ZHAO L, CHEN Y Z, et al. Eye-gaze-guided
[107] ZHENG Z, WEI J, XU Y R, et al. Modeling
vision transformer for rectifying shortcut learning[J].
IEEE Trans Med Imaging, 2023, 42(11): 3384-3394. macroscopic brain dynamics with brain-inspired
[94] WANG S, ZHAO Z H, SHEN Z R, et al. Improving computing architecture[J]. Nat Commun, 2025, 16(1):
self-supervised medical image pre-training by early 9424.
alignment with human eye gaze information[J]. IEEE [108] PEI J, DENG L, SONG S, et al. Towards artificial
Trans Med Imaging, 2025, 44(10): 4063-4072. general intelligence with hybrid Tianjic chip
[95] LALONDE R, XU Z Y, IRMAKCI I, et al. Capsules for architecture[J]. Nature, 2019, 572(7767): 106-111.
biomedical image segmentation[J]. Med Image Anal, [109] MEROLLA P A, ARTHUR J V, ALVAREZ-ICAZA R,
2021, 68: 101889. et al. Artificial brains. A million spiking-neuron
[96] CHEN J, LU Y, YU Q, et al. TransUNet: transformers integrated circuit with a scalable communication
make strong encoders for medical image segmentation network and interface[J]. Science, 2014, 345(6197):
[EB/OL]. arXiv preprint, 2021[2025-11-05]. https:// 668-673.
180