Objective  To investigate the value of ultrasound-based intratumoral and peritumoral radiomics in predicting lymph node metastasis from primary thyroid cancer.

Methods  A retrospective collection of patient data from January 2016 to April 2023 at the Sixth People’s Hospital (Lingang Campus) in Shanghai was conducted for patients who underwent routine ultrasound examinations and were pathologically diagnosed with thyroid cancer. On the two-dimensional ultrasound images, the largest cross-sectional view of the lesion was selected to delineate the region of interest. This region was then automatically expanded by 2 mm outward to define the peritumoral area. Intratumoral and peritumoral radiomics were used to extract features. Pationts were randomly divided into training group and validation group in a 7 : 3 ratio. The least absolute shrinkage and selection operator algorithm was employed to select radiomic features. Multiple layer perceptron models were used to establish several predictive models: a clinical variable model, an intratumoral radiomics model, a peritumoral radiomics model, a combined intratumoral and peritumoral radiomics model, and a model that integrates radiomics features with clinical variables. Receiver operating characteristic curves and area under curve (AUC) were used to evaluate the model's prediction performance for lymph node metastasis.

Results  Among 280 cases of thyroid cancer, 102 cases (36.43%) had lymph node metastasis following surgical treatment. The patients were randomly divided into a training group of 196 cases and a validation group of 84 cases. Multivariate Logistic regression analysis reveals a notable correlation among patient age, thyroid peroxidase antibody, and the proximity of nodules to the thyroid capsule, with the occurrence of lymph node metastasis in thyroid cancer (P<0.05). The AUC values for the validation group for the clinical variable model, the intratumoral ultrasound radiomics model, the peritumoral ultrasound radiomics model, the combined intratumoral and peritumoral ultrasound radiomics model, and the model integrating intratumoral and peritumoral ultrasound radiomics features with clinical variables were 0.651[95%CI(0.527, 0.775)], 0.687[95%CI(0.570, 0.803)], 0.696[95%CI(0.575, 0.817)], 0.737[95%CI(0.629, 0.845)] and 0.738[95%CI(0.629, 0.847)]. The diagnostic efficacy of the combined intra- and peritumoral radiomics model with the clinical variables was better than that of the clinical variables group and intratumoral ultrasound-based radiomics, with statistically significant differences (P<0.05). It was higher than the peritumoral and intratumoral combined with peritumoral models, but there was no significant difference (P>0.05).

Conclusion  Intratumor and peritumor ultrasonographic radiomics have certain value in predicting lymph node metastasis of thyroid cancer. The combined features of intratumor and peritumor ultrasonographic radiomics can enhance the prediction accuracy for lymph node metastasis in thyroid cancer, providing a reference for clinical diagnosis and treatment.

1.Xi X, Wang Y, Gao L, et al. Establishment of an ultrasound malignancy risk stratification model for thyroid nodules larger than 4 cm[J]. Front Oncol, 2021, 11: 592927. DOI: 10.3389/FONC. 2021.592927.

2.李朋禺,赵婉君,李志辉,等. 甲状腺癌手术入院优先级模型的构建方法[J]. 中国循证医学杂志, 2022, 22(6): 628-633. [Li PY, Zhao WJ, Li ZH, et al. Method for constructing a priority model for surgical admission for thyroid cancer[J]. Chinese Journal of Evidence-Based Medicine, 2022, 22(6): 628-633.] DOI: 10.7507/1672-2531.202112112.

3.Lee JY, Yoo RE, Rhim JH, et al. Validation of ultrasound risk stratification systems for cervical lymph node metastasis in patients with thyroid cancer[J]. Cancers (Basel), 2022, 14(9): 2106. DOI: 10.3390/CANCERS14092106.

4.Gao L, Li X, Xia Y, et al. Large-Volume lateral lymph node metastasis predicts worse prognosis in papillary thyroid carcinoma patients with N1b[J]. Front Endocrinol (Lausanne), 2022, 12: 815207. DOI: 10.3389/FENDO.2021.815207.

5.Xue T, Liu C, Liu JJ, et al. Analysis of the relevance of the ultrasonographic features of papillary thyroid carcinoma and cervical lymph node metastasis on conventional and contrast-enhanced ultrasonography[J]. Front Oncol, 2021, 11: 794399. DOI: 10.3389/FONC.2021.794399.

6.Zhou TH, Zhao LQ, Zhang Y, et al. The prediction of metastases of lateral cervical lymph node in medullary thyroid carcinoma[J]. Front Endocrinol (Lausanne), 2021, 12: 741289. DOI: 10.3389/FENDO.2021.741289.

7.Yoo RE, Kim JH, Hwang I, et al. Added value of computed tomography to ultrasonography for assessing LN metastasis in preoperative patients with thyroid cancer: node-by-node correlation[J]. Cancers (Basel), 2020, 12(5): 1190. DOI: 10.3390/cancers 12051190.

8.邓红梅,毛玲玲,钟青玉,等. 增强CT联合高分辨率超声诊断甲状腺癌颈部淋巴结转移的临床价值研究[J]. 中国医学装备, 2023, 20(8): 49-52. [Deng HM, Mao  LL, Zhong  QY, et al. Clinical value of contrast-enhanced CT combined with high-resolution ultrasonography in the diagnosis of cervical lymph node metastasis of thyroid cancer[J]. China Medical Equipment, 2023, 20(8): 49-52.] DOI: 10.3969/J.ISSN.1672-8270. 2023.08.010.

9.Turk AT, Asa SL, Baloch ZW, et al. Interobserver variability in the histopathologic assessment of extrathyroidal extension of well differentiated thyroid carcinoma supports the new American joint committee on cancer eighth edition criteria for tumor staging[J]. Thyroid, 2019, 29(5): 619-624. DOI: 10.1089/thy.2018.0286.

10.Minna E, Brich S, Todoerti K, et al. Cancer associated fibroblasts and senescent thyroid cells in the invasive front of thyroid carcinoma[J]. Cancers (Basel), 2020, 12(1): 112. DOI: 10.3390/cancers12010112.

11.Zhao J, Zhou X, Shi G, et al. Semantic consistency generative adversarial network for cross-modality domain adaptation in ultrasound thyroid nodule classification[J]. Appl Intell (Dordr), 2022, 52(9): 10369-10383. DOI: 10.1007/S10489-021-03025-7.

12.Avanzo M, Stancanello J, El Naqa I. Beyond imaging: the promise of radiomics[J]. Phys Med, 2017, 38: 122-139. DOI: 10.1016/j.ejmp.2017.05.071.

13.Tibshirani R. Regression shrinkage and selection via the lasso: a retrospective[J]. J R Stat Soc Series B Stat Methodol, 2011, 73(3): 273-282. DOI: 10.1111/j.1467-9868.2011.00771.x.

14.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019[J]. CA Cancer J Clin, 2019, 69(1): 7-34. DOI: 10.3322/caac.21551.

15.Huang C, Hu D, Zhuang Y, et al. Risk factors and prediction model of level II lymph node metastasis in papillary thyroid carcinoma[J]. Front Oncol, 2022, 12: 984038. DOI: 10.3389/FONC.2022.984038.

16.Wang Y, Deng C, Shu X, et al. Risk factors and a prediction model of lateral lymph node metastasis in CN0 papillary thyroid carcinoma patients with 1-2 central lymph node metastases[J]. Front Endocrinol (Lausanne), 2021, 12: 716728. DOI: 10.3389/FENDO.2021.716728.

17.Jin P, Chen J, Dong Y, et al. Ultrasound-based radiomics nomogram combined with clinical features for the prediction of central lymph node metastasis in papillary thyroid carcinoma patients with Hashimoto's thyroiditis[J]. Front Endocrinol (Lausanne), 2022, 13: 993564. DOI: 10.3389/FENDO.2022. 993564.

18.张瑞坚,刘立衡,刘巧爱,等. 术前超声特征诊断甲状腺乳头状癌颈部淋巴结转移的价值[J]. 中国现代普通外科进展, 2021, 24(3): 217-219. [Zhang RJ, Liu LH, Liu QA, et al. The value of preoperative ultrasound features in diagnosing cervical lymph node metastasis in papillary thyroid carcinoma[J]. Chinese Journal of Current Advances in General Surgery, 2021, 24(3): 217-219.] DOI: 10.3969/j.issn.1009-9905.2021.03.012.

19.Xu H, Wang X, Guan C, et al. Value of whole-thyroid CT-based radiomics in predicting benign and malignant thyroid nodules[J]. Front Oncol, 2022, 12: 828259. DOI: 10.3389/FONC. 2022.828259.

20.周世崇,刘桐桐,周瑾,等. 影像组学在甲状腺癌应用的初步研究[J]. 肿瘤影像学, 2017, 26(2): 102-105. [Zhou SC, Liu  TT, Zhou J, et al. Preliminary study on application of radiomics in thyroid carcinoma[J]. Oncoradiology, 2017, 26(2): 102-105.] DOI: 10.3969/j.issn.1008-617X.2017.02.004.

21.Liu T, Zhou S, Yu J, et al. Prediction of lymph node metastasis in patients with papillary thyroid carcinoma: a radiomics method based on preoperative ultrasound images[J]. Technol Cancer Res Treat, 2019, 18: 1533033819831713. DOI: 10.1177/ 1533033819831713.

22.Ding J, Chen S, Serrano Sosa M, et al. Optimizing the peritumoral region size in radiomics analysis for sentinel lymph node status prediction in breast cancer[J]. Acad Radiol, 2020, 29(Suppl 1): S223-S228. DOI: 10.1016/j.acra.2020.10.015.

23.Kawazoe Y, Shiinoki T, Fujimoto K, et al. Investigation of the combination of intratumoral and peritumoral radiomic signatures for predicting epidermal growth factor receptor mutation in lung adenocarcinoma[J]. J Appl Clin Med Phys, 2023, 24(6): e13980. DOI: 10.1002/ACM2.13980.

24.Vural Topuz Ö, Aksu A, Yılmaz Özgüven MB. A different perspective on 18F-FDG PET radiomics in colorectal cancer patients: the relationship between intra & peritumoral analysis and pathological findings[J]. Rev Esp Med Nucl Imagen Mol (Engl Ed), 2023, 42(6): 359-366. DOI: 10.1016/J.REMNIE.2023.04.005.