Page 38 - 《中国医疗器械杂志》2026年第1期

Chinese Journal of Medical Instrumentation 2026年 第50卷 第1期

医 学 人 工 智 能

survival of patients with metachronous or synchronous shear-wave elastography and color Doppler: feasible
double primary malignancies: breast and thyroid method to avoid unnecessary breast excision of
cancer[J]. Oncotarget, 2016, 7(32): 52450-52459. fibroepithelial lesions diagnosed by core needle biopsy[J].
[18] LIU Y, ZHAO H, HUANG J, et al. A deep learning PLoS One, 2017, 12(5): e0175380.
system for thyroid nodule diagnosis and classification [28] KIM Y J, CHOI Y, HUR S J, et al. Deep convolutional
from ultrasound images[J]. Nat Commun, 2022, 13(1): neural network for classification of thyroid nodules on
6457. ultrasound: comparison of the diagnostic performance
[19] WANG J, LIU W, NING Z, et al. An intelligent diagnosis with that of radiologists[J]. Eur J Radiol, 2022, 152:
method of breast and thyroid nodules based on neural 110335.
architecture search of ultrasonic multimodal images[J]. [29] DENG M, CHEN Z, GAO Y, et al. Optimal ultrasound
Front Oncol, 2021, 11: 743047.
feature ranking for malignancy prediction based on
[20] SILVA J M, SALDANHA C, MARTINS E SILVA J.
machine learning in thyroid nodules[J]. Eur Radiol, 2023,
Hormones, growth factors, and breast cancer: basic and
33(1): 458-467.
clinical approach[J]. Front Med (Lausanne), 2021, 8:
[30] SCHAFFTER T, BUIST D S M, LEE C I, et al.
745308.
Evaluation of combined artificial intelligence and
[21] PASCHOU S A, VRYONIDOU A, GOULIS D G.
radiologist assessment to interpret screening
Thyroid nodules: a guide to assessment, treatment and
mammograms[J]. JAMA Netw Open, 2020, 3(3):
follow-up[J]. Maturitas, 2017, 96: 1-9.
e200265.
[22] JOSEPH K R, EDIRIMANNE S, ESLICK G D. The
[31] LIM H, DEVESA S S, SOSA J A, et al. Trends in thyroid
association between breast cancer and thyroid cancer: a
cancer incidence and mortality in the United States, 1974-
meta-analysis[J]. Breast Cancer Res Treat, 2015, 152(1):
2013[J]. JAMA, 2017, 317(13): 1338-1348.
173-181.
[32] DESANTIS C E, MA J, GAUDET M M, et al. Breast
[23] ZHANG Y, YANG Q. A survey on multi-task
cancer statistics, 2019[J]. CA Cancer J Clin, 2019, 69(6):
learning[J]. IEEE Trans Knowl Data Eng, 2022, 34(12):
438-451.
5586-5609.
[33] BENJAMENS S, DHUNNOO P, MESKÓ B. The state of
[24] FAN Y H, WALDMANN P. Tabular deep learning: a
artificial intelligence-based FDA-approved medical
comparative study applied to multi-task genome-wide
prediction[J]. BMC Bioinformatics, 2024, 25(1): 322. devices and algorithms: an online database[J]. NPJ Digit
[25] MINAGI A, HIRANO H, TAKEMOTO K. Natural Med, 2020, 3: 118.
images allow universal adversarial attacks on medical [34] LEE G, KANG B, NHO K, et al. MildNet: a lightweight
image classification using deep neural networks with single-scaled deep learning model for real-time thyroid
transfer learning[J]. J Imaging, 2022, 8(2): 38. nodule detection in ultrasound videos[J]. Med Image
[26] NIELSEN S M, WHITE M G, HONG S, et al. The Anal, 2021, 68: 101909.
breast-thyroid cancer link: a systematic review and meta- [35] ZHOU Z W, SODHA V, RAHMAN SIDDIQUEE M M,
analysis[J]. Cancer Epidemiol Biomarkers Prev, 2016, et al. Models Genesis: generic autodidactic models for 3D
25(2): 231-238. medical image analysis[J]. Med Image Comput Comput
[27] KIM G R, CHOI J S, HAN B K, et al. Combination of Assist Interv, 2019, 11767: 384-393.

34