Page 63 - 《应用声学》2021年第1期

第 40 卷 第 1 期 高凡等： 超声射频信号的甲状腺结节智能诊断方法 59

衷，有望在此基础上建立一套可用于甲状腺结节的 [12] Fang D, Ma W, Xu L, et al. A predictive model to dis-
预筛查的临床智能诊断系统。 tinguish papillary thyroid carcinomas from benign thyroid
nodules using ultrasonographic features: a single-center,
retrospective analysis[J]. Medical Science Monitor, 2019,
25: 9409–9415.
参 考 文 献
[13] Yin L, Zhang W, Bai W, et al. Relationship between
morphologic characteristics of ultrasonic calcification in
[1] Chen W, Zheng R, Baade P D, et al. Cancer statistics thyroid nodules and thyroid carcinoma[J]. Ultrasound in
in China, 2015[J]. CA: A Cancer Journal for Clinicians, Medicine and Biology, 2020, 46(1): 20–25.
2016, 66(2): 115–132. [14] Hu L, He N, Ye L, et al. Evaluation of the stiffness of tis-
[2] Liu Z, Jiang Y, Fang Q, et al. Future of cancer incidence sues surrounding thyroid nodules with shear wave elastog-
in Shanghai, China: predicting the burden upon the age- raphy[J]. Journal of Ultrasound in Medicine, 2018, 37(9):
ing population[J]. Cancer Epidemiol, 2019, 60: 8–15. 2251–2261.
[3] Nikiforov Y E, Ohori N P, Hodak S P, et al. Impact [15] Sun C, Zhang Y, Chang Q, et al. Evaluation of a
of mutational testing on the diagnosis and management deep learning-based computer-aided diagnosis system for
of patients with cytologically indeterminate thyroid nod- distinguishing benign from malignant thyroid nodules
ules: a prospective analysis of 1056 FNA samples[J]. Jour- in ultrasound images[J]. Medical Physics, 2020, 47(9):
nal of Clinical Endocrinology & Metabolism, 2011, 96(11): 3952–3960.
3390–3397. [16] Liang X, Yu J, Liao J, et al. Convolutional neural net-
[4] Moreira-Souza L, Michels M, de Melo L P L, et al. Bright- work for breast and thyroid nodules diagnosis in ultra-
ness and contrast adjustments influence the radiographic sound imaging[J]. BioMed Research International, 2020,
detection of soft tissue calcification[J]. Oral Disease, 2019, 2020: 1763803.
25(7): 1809–1814. [17] Ma J, Wu F, Zhu J, et al. A pre-trained convolutional
[5] Khasawneh A, Takeshita Y, Hisatomi M, et al. Incidental neural network based method for thyroid nodule diagno-
findings in the thyroid gland on computed tomography sis[J]. Ultrasonics, 2017, 73: 221–230.
images of the oral and maxillofacial region[J]. Oncology [18] Liu C, Xie L, Kong W, et al. Prediction of suspicious
Letters, 2020, 19(3): 2005–2010. thyroid nodule using artificial neural network based on
[6] Sarlis J N, Brucker-Davis F, Doppman J L, et al. MRI- radiofrequency ultrasound and conventional ultrasound:
demonstrable regression of a pituitary mass in a case of a preliminary study[J]. Ultrasonics, 2019, 99: 105951.
primary hypothyroidism after a week of acute thyroid hor- [19] Mishra V, Rath S K. Detection of breast cancer tumours
mone therapy[J]. Journal of Clinical Endocrinology and based on feature reduction and classification of thermo-
Metabolism, 1997, 82(3): 808–811. grams[J]. Quantitative Infrared Thermography Journal,
[7] Acharya U R, Sree V S, Krishnan M R M, et al. Non- 2020: 1–14.
invasive automated 3D thyroid lesion classification in ul- [20] Tsui P H, Wan Y L. Effects of fatty infiltration of the
trasound: a class of ThyroScan systems[J]. Ultrasonics, liver on the Shannon entropy of ultrasound backscattered
2012, 52(4): 508–520. signals[J]. Entropy, 2016, 18(9): 341.
[8] Tessler F N, Middleton W D, Grant E G, et al. ACR [21] Tsui P H. Ultrasound detection of scatterer concen-
thyroid imaging, reporting and data system (TI-RADS): tration by weighted entropy[J]. Entropy, 2015, 17(10):
white paper of the ACR TI-RADS committee[J]. Jour- 6598–6616.
nal of the American College of Radiology, 2017, 14(5): [22] Ma H Y, Zhou Z, Wu S, et al. A computer-aided diagnosis
587–595. scheme for detection of fatty liver in vivo based on ultra-
[9] Xu H, Liu C, Yang P, et al. A nonlinear approach to sound kurtosis imaging[J]. Journal of Medical Systems,
identify pathological change of thyroid nodules based on 2016, 40(1): 33.
statistical analysis of ultrasound RF signals[J]. Scientific [23] Tsui P H, Huang C C, Chang C C, et al. Feasibility
Reports, 2017, 7(1): 16930. study of using high-frequency ultrasonic Nakagami imag-
[10] Rohrbach D, Smith J, Goundan P, et al. Quantita- ing for characterizing the cataract lens in vitro[J]. Physics
tive ultrasound-based detection of cancerous thyroid nod- in Medicine and Biology, 2007, 52(21): 6413–6425.
ules[C]. 2018 IEEE International Ultrasonics Symposium [24] Zhu L C, Ye Y L, Luo W H, et al. A model to discrimi-
(IUS), 2018: 1–9. nate malignant from benign thyroid nodules using artifi-
[11] Shi Y Z, Jin Y, Zheng L. Partially cystic thyroid nodules cial neural network[J]. PLoS One, 2013, 8(12): e82211.
on ultrasound: the associated factors for malignancy[J]. [25] Hill T, Marquez L, O’Connor M, et al. Artificial neural
Clinical Hemorheology and Microcirculation, 2020, 74(4): network models for forecasting and decision making[J].
373–381. International Journal of Forcasting, 1994, 10(1): 5–15.