
==== Front
Insights Imaging
Insights Imaging
Insights into Imaging
1869-4101
Springer Vienna Vienna

39311997
1806
10.1186/s13244-024-01806-5
Original Article
Application of microvascular ultrasound-assisted thyroid imaging report and data system in thyroid nodule risk stratification
Ma Guangrong 12
Chen Libin 3
Wang Yong 4
Luo Zhiyan 1
Zeng Yiqing 12
Wang Xue 12
Shi Zhan 12
http://orcid.org/0000-0003-4968-9120
Zhang Tao zhangtao-us@zju.edu.cn

12
Hong Yurong hongyurong@zju.edu.cn

12
Huang Pintong huangpintong@zju.edu.cn

125
1 https://ror.org/059cjpv64 grid.412465.0 Department of Ultrasound in Medicine, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, P.R. China
2 grid.13402.34 0000 0004 1759 700X Research Center of Ultrasound in Medicine and Biomedical Engineering, The Second Affiliated Hospital of Zhejiang University School of Medicine, Zhejiang University, Hangzhou, P.R. China
3 grid.460077.2 0000 0004 1808 3393 Department of Ultrasound in Medicine, The First Affiliated Hospital of Ningbo University, Ningbo, P.R. China
4 grid.13402.34 0000 0004 1759 700X Department of Thyroid Surgery, The Second Affiliated Hospital of Zhejiang University School of Medicine, Zhejiang University, Hangzhou, P.R. China
5 https://ror.org/00a2xv884 grid.13402.34 0000 0004 1759 700X Research Center for Life Science and Human Health, Binjiang Institute of Zhejiang University, Hangzhou, P.R. China
23 9 2024
23 9 2024
12 2024
15 23029 4 2024
29 8 2024
© The Author(s) 2024
2024
https://creativecommons.org/licenses/by/4.0/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Objectives

To establish superb microvascular imaging (SMI) based thyroid imaging reporting and data system (SMI TI-RADS) for risk stratification of malignancy in thyroid nodules.

Methods

In total, 471 patients, comprising 643 thyroid nodules, who received conventional ultrasound (US), SMI, and a final diagnosis were extensively analyzed. A qualitative assessment of US features of the nodules was performed followed by univariable and multivariable logistic regression analyses, leading to the construction of the SMI TI-RADS, which was further verified using internal and external validation cohorts.

Results

Among the stand-alone US, predictive factors were the shape and margins of the nodules, echogenicity and echogenic foci, vascularity, extrathyroidal extension, ring-SMI patterns, penetrating vascularity, flow-signal enlarged, and vascularity area ratio. SMI TI-RADS depicted an enhanced area under the receiver operating characteristic curve (AUC) of 0.94 (95% CI: 0.92, 0.96; p < 0.001 relative to other stratification systems), a 79% biopsy yield of malignancy (BYM, 189/240 nodules), and a 21% unnecessary biopsy rate (UBR, 51/240 nodules). In the verification cohorts, we demonstrated AUCs, malignancy biopsy yields, and unnecessary biopsy rates of 0.88 (95% CI: 0.83, 0.94), 79% (59/75 nodules), and 21% (16/75 nodules) for the internal cohort, respectively, and 0.91 (95% CI: 0.85, 0.96), 72% (31/43 nodules), and 28% (12/43 nodules) for the external cohort, respectively.

Conclusion

SMI TI-RADS was found to be superior in diagnostic sensitivity, specificity, and efficiency than existing TI-RADSs, showing better stratification of the malignancy risk, and thus decreasing the rate of unnecessary needle biopsy.

Critical relevance statement

To develop an imaging and data system based on conventional US and SMI features for stratifying the malignancy risk in thyroid nodules.

Key Points

SMI features could improve thyroid nodule risk stratification.

SMI TI-RADS showed superior diagnostic efficiency and accuracy for biopsy guidance.

SMI TI-RADS can provide better guidance for clinical diagnosis and treatment of thyroid nodules.

Graphical Abstract

Keywords

Thyroid
Thyroid cancer
Ultrasound imaging
https://doi.org/10.13039/http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 82030048 82030048 82030048 82030048 82030048 82030048 82030048 Chen Libin Wang Xue Shi Zhan Zhang Tao Huang Pintong https://doi.org/10.13039/http://dx.doi.org/10.13039/501100004731 Natural Science Foundation of Zhejiang Province LQ24H180004 Shi Zhan https://doi.org/10.13039/ The General Research Program of Health Commission in Zhejiang Province 2024KY1089 Luo Zhiyan issue-copyright-statement© European Society of Radiology (ESR) 2024
==== Body
pmcIntroduction

The incidence of thyroid cancer is increasing [1–3]. Ultrasound (US) has shown that between 19% and 68% of the overall population possess thyroid nodules, the majority of which are benign and clinically insignificant, with only 7–15% showing evidence of malignancy [4, 5]. Small malignant tumors usually have a slow course and usually do not cause significant complications or death during the patient’s lifetime, so their detection may constitute overdiagnosis [6, 7]. Fine-needle aspiration (FNA) is both cost-effective and accurate for thyroid nodule evaluation; however, many nodules require neither biopsy nor surgery [4, 8–10]. Thus, stratification of the risk of malignancy is desirable, and the use of risk stratification systems (RSSs) based on the US have been developed for the accurate assessment of the malignancy risk in thyroid modules and thus reduce unnecessary biopsies.

There are currently no fewer than 20 RSSs in use. These include pattern- and non-pattern-based systems, based on the combination of nodule size and conventional US features to provide guidance on the need for FNA biopsy [11–15]. However, most of these RSS do not analyze the vascular distribution and blood flow, both of which are closely related to the biological characteristics of tumors, considering the limitations of vascular characteristic imaging modalities [16, 17]. Examples of these limitations include the technical limitations of color Doppler flow imaging (CDFI) in the detection of small vessels and reduced blood flow due to its imaging principles, and limitations in the application of contrast-enhanced ultrasound imaging in terms of the contrast agent injection and economic conditions [18, 19].

Microvascular ultrasound (MVUS), is a form of Doppler imaging mode that can visualize low-speed blood flow and thus provide a more detailed analysis of vascular blood flow than CDFI, leading to the acquisition of high-quality images of blood flow without the use of contrast agents [20–22]. A meta-analysis found that the sensitivity, specificity, and positive and negative likelihood ratios of SMI (superb microvascular imaging, MVUS based on Canon Medical System) for diagnosing malignant thyroid nodules were 0.84, 0.86, 6.2, and 0.18, respectively, which were all better than CDFI [20, 23].

Here, we propose to use conventional US and MVUS (based on Canon Medical System) to create an easily generalizable and practical risk stratification system for thyroid nodule malignancy, termed SMI thyroid imaging reporting and data system (TI-RADS).

Materials and methods

Patient population and data collection

This research received ethical approval from the Second Affiliated Hospital of Zhejiang University School of Medicine (no. 2023-0944), and the participant informed consent requirement was waived as the study was retrospective. Regardless, all subjects who received FNA or surgery provided informed consent prior to the procedures. The inclusion criteria were: (a) those who underwent conventional US, SMI, and FNA in our department between January 2020 and December 2021; (b) those with nodules > 5 mm in maximum diameter, as detected by routine US; (c) those with an interval not more than 2 weeks between FNA and surgery, during which time, they received no clinical intervention; and (d) those with pathologically confirmed benign or malignant nodules. Additionally, we eliminated the following patients from the analysis: (a) those with nodules stratified as Bethesda I, III, or IV based on the Bethesda Thyroid Cell Pathology Reporting System, with no evident pathological reports; (b) those with prior FNA or ablation; and (c) those with poor image quality, with severe artifacts or low resolution. The training and internal validation cohorts were recruited from the Second Affiliated Hospital of Zhejiang University School of Medicine, and the external validation cohort was obtained from the First Affiliated Hospital of Ningbo University.

Pathological reference standards

The pathological reference standards were defined as follows: (a) surgical resection, as evidenced by histology; (b) the Bethesda system-based classification of malignant results (V, VI) without histopathological assessment; and (c) employment of the Bethesda system (II), for benign thyroid nodule identification without histopathological assessment, with the depiction of nodule stability or reduced size seen on the US over a 12-month follow-up period [24].

US protocol

US-based assessments utilized a Canon Aplio i900 (Canon Medical System Corporation, Canon, Japan) equipped with an i18LX5 high-frequency (8–15 MHz) line array transducers, with conventional US and SMI functions. All participants were initially evaluated using conventional US, using parameters as follows: depth of 2.5–3 cm, with a focus on the middle to rear edge of the nodule, and dynamic range of 65 dB. Upon detection of a thyroid nodule, we recorded its size, position, and other US features, such as, composition, echogenicity, shape, margin, echogenic foci, halo, and extrathyroidal extension condition. Subsequently, we employed SMI to assess the vascular profile and morphology using parameters as follows: velocity scale, 1–1.5 cm/s; frame rate, 59 fps; and color gain, 41 dB. During the US procedure, patients were recommended to avoid swallowing and to breathe slowly to minimize motion. We also applied slight pressure through the transducer during imaging to avoid vessel collapse. Using the aforementioned process, we evaluated the vasculature in and around the lesions.

Statistical analysis

We based our participant recruitment on 31% carcinoma incidence according to the ACR TI-RADS study, as well as an estimate of 85% sensitivity for the RSS for target lesions [25]. We attempted to include at least 561 thyroid nodules to yield a 91% power for 75–85% sensitivity detection using a two-sided binomial test. Continuous data were assessed via the unpaired t-test, whereas, categorical data utilized the χ2 test. Univariable logistic regression analysis was employed for the determination of significant US features (p value < 0.05), which were subsequently entered into the multivariable analysis. Thereafter, we rounded the average regression coefficients from multivariable analysis following 10-fold cross-validation to obtain scores for SMI TI-RADS generation, which were modeled after the ACR TI-RADS, using previously described methods. SMI TI-RADS, along with eight representative classification systems, namely, C TI-RADS, ACR TI-RADS, European Thyroid Association system (EU TI-RADS), Korean Society of Thyroid Radiology Imaging Guidelines (KSThR TI-RADS), American Association of Clinical Endocrinologists system (AACE TI-RADS), American Thyroid Association Management Guidelines (ATA TI-RADS), British Thyroid Association system (BTA TI-RADS), and French Society of Endocrinology system (FSE TI-RADS), were then employed for thyroid nodule evaluation. To elucidate each stratification system performance for patient biopsy requirement, we computed the biopsy yield of malignancy (BYM, representing the percentage of malignant nodules in the overall number of nodules required for biopsy) and the unnecessary biopsy rate (UBR, representing the percentage of benign nodules in the overall number of nodules required for biopsy), as described in a prior investigation, nodules with indications for biopsy were considered positive and those without indications for biopsy were considered negative. We next compared the SMI TI-RADS diagnostic and biopsy performances against the eight aforementioned classification systems.

Data analyses utilized the R-Project and R Studio (version 4.3.1), and two-tailed p < 0.05 was regarded as significant.

Results

Baseline demographics of study participants

Table 1 details the baseline profiles of all patients. Our initial sample size included 522 patients with 699 nodules, recruited from patients who received conventional US, SMI mode US, and FNA at our institution between January 2020 and December 2021. Of these, 5 nodules were eliminated owing to a prior history of FNA (n = 2) or ablation (n = 3). Additionally, 56 nodules, particularly, 39 Bethesda I, 15 Bethesda III, and 7 Bethesda IV nodules, were eliminated due to unavailable pathological reference standards (Fig. 1). In the final training cohort, there were 471 participants (median age, 42 years [IQR, 34–54 years]), among which 103 were males (median age, 41 years [IQR, 34–53 years]) and 368 were females (median age, 42 years [IQR, 34–54 years]). Moreover, the total nodules were 643 (median size 10.9 mm, 7.5–13.9 mm), among which, 298 were benign thyroid nodules (median size 10.8 mm, [IQR, 7.2–15.8 mm]) and 345 were malignant thyroid nodules (median size 10.9 mm, [IQR, 7.9–12.8 mm]).Table 1 Baseline data of training set and validation sets

Parameter	Training set	p value	Internal validation cohort	p value	External validation cohort	p value	
Number of nodules	643		159		111		
Sex&	
 F	510 (79.3)		133 (83.6)		86 (77.5)		
 M	133 (20.7)		26 (16.4)		25 (22.5)		
Age (y)*		0.76		0.68		0.43	
 F	42 (34, 54)		43 (34, 56)		48.5 (36, 56)		
 M	41 (34, 53)		40 (35, 48.5)		42 (34, 49)		
Nodules&	
 Benign	298 (46.3)		58 (36.5)		50 (45.0)		
 Malignant	345 (53.7)		101 (63.5)		61 (55.0)		
Size*		0.070		0.213		< 0.001	
 Benign	10.80 (7.23, 15.80)		10.90 (7.20, 16.58)		11.95 (10.22, 16.75)		
 Malignant	10.90 (7.90, 12.80)		10.60 (7.40, 13.10)		10.20 (7.10, 12.40)		
Position&	
 Left	287 (44.6)		75 (47.2)		49 (44.1)		
 Right	320 (49.8)		74 (46.5)		55 (49.5)		
 Isthmus	36 (5.6)		14 (8.8)		7 (6.3)		
* Median with 95% CI

& Data are numbers of nodules, with percentages in parentheses

Fig. 1 Flow diagram of the included patients and number of thyroid nodules. SMI, superb microvascular imaging; FNA, fine-needle aspiration; n, number of thyroid nodules

Between the period of January 2022 and June 2022, 159 thyroid nodules from 144 patients (median age, 42 years [IQR, 34–56 years]) were selected from our internal validation cohort, among which, 120 were females (median age, 43 years, [IQR, 34–56 years]) and 24 were males (median age, 40 years, [IQR, 35–49 years]). Additionally, between January 2021 and December 2021, as our external validation cohort, we selected 111 thyroid nodules from 106 patients (median age, 46 years [IQR, 36–56 years], among which, 81 were females (median age, 49 years, [IQR, 36–56 years]) and 25 were males (median age, 42 years, [IQR, 34–49 years]). Table 1 summarizes the baseline profiles of internal and external validation cohorts.

Development of the SMI TI-RADS estimation model

According to ACR TI-RADS, we established SMI TI-RADS for thyroid nodule malignancy prediction and FNA indication assessment. The univariable (Table S1 [online]) and multivariable logistic regression analyses indicated that nodule shape, margins, echogenicity, echogenic foci, extrathyroidal extension, vascularity, ring-SMI pattern, penetrating vascularity, flow-signal enlarged were strong indicators of thyroid carcinoma, and therefore, these parameters were employed for establishment of the SMI TI-RADS estimation model (Tables 2 and S2 [online]). Within this model, hyperechoic or isoechoic, shapes that were wider rather than tall, smooth margins, a lack of calcification or large comet tail artifacts, absence of an extrathyroidal extension, absent or mixed type of vascularity, smooth ring-SMI pattern, none too rare penetrating vascularity, and absent of flow-signal enlarged were given 0 points, while nodules that were either hypoechoic or highly hypoechoic, margins that were lobulated or irregular, peripheral or macrocalcifications, extrathyroidal extensions, central or perinodular type vascularity, absent of ring-SMI patterns, and present of flow-signal enlarged were given 1 point, and punctate echogenic foci, showing greater height than width, punctate echogenic foci, multiple penetrating vascularity, and cloud ring-SMI pattern were given 2 points (Figs. 2–5 and S1 [online]).Table 2 Association between thyroid nodule malignancy and various SMI features

	Benign	Malignant	Univariable analysis	Multivariable analysis		
Features	(n = 298)	(n = 345)	β	p value#	β&	p value#	Score▲	
Vascularity				< 0.001				
None/mixed	218	143	N/A		N/A	N/A	0	
Perinodular/central	80	202	1.35		1.44 (1.38, 1.49)	< 0.001	1	
Vascularity area ratio				0.475				
 < 1/3	151	164	N/A		N/A	N/A	N/A	
 ≥ 1/3	147	181	0.13		N/A	N/A	N/A	
Ring-SMI patterns				< 0.001				
 Smooth	216	115	N/A		N/A	N/A	0	
 None	44	51	0.78		1.21 (1.12, 1.29)	< 0.001	1	
 Cloud	38	179	2.18		2.30 (2.24, 2.36)	< 0.001	2	
Penetrating vascularity				< 0.001				
 None	249	135	N/A		N/A	N/A	0	
 Rare	26	19	0.30		0.53 (0.42, 0.63)	0.158	0	
 Multiple	23	191	2.73		2.40 (2.33, 2.47)	< 0.001	2	
Flow-signal enlarged				< 0.001				
 None	256	189	N/A		N/A	N/A	0	
 Expansion	42	156	1.62		1.31 (1.27, 1.35)	< 0.001	1	
# Determined with logistic regression analysis

& Mean and 95% CI of regression coefficients of significant predictive both US and SMI features after 10-fold cross-validation

▲ Scoring criteria for significant independent predictors were based on the severely rounded mean of regression coefficients after 10-fold cross-validation to the nearest integer

Fig. 2 Monochrome SMI and conventional US images show features of vascularity

Fig. 3 Monochrome SMI and conventional US images show features of ring-SMI patterns

Fig. 4 Monochrome SMI and conventional US images show features of penetrating vascularity

Fig. 5 Monochrome SMI and conventional US images show features of the area enlarged

The total points determined the nodule SMI TI-RADS category (referred to as TR hereafter): 0 points represented TR1 (benign thyroid nodules; fitted probabilities [FP]: 0%), 1 point indicated TR2 (not suspicious; FP: 1%), 2 points meant TR3 (mildly suspicious; FP: 2–5%), 3 points represented TR4A (moderately suspicious; FP: 9–17%), 4 points indicated TR4B (moderately suspicious; FP: 24–44%), 5 points meant TR4C (moderately suspicious; FP: 56–75%), and 6–11 points represented TR5 (highly suspicious; FP: > 90%) (Fig. S2 [online]). The indication for FNA was dependent on an integration of nodule maximum diameter and classification (Fig. 6).Fig. 6 Chart shows the five categories of the SMI TI-RADS, with FP and indications for FNA

Comparison between SMI TI-RADS and other risk classification systems

In all, we employed nine RSSs (namely, C TI-RADS, ACR TI-RADS, EU TI-RADS, KSThR TI-RADS, AACE TI-RADS, ATA TI-RADS, and SMI TI-RADS) to assess the risk of malignancy in the classifications of these guidelines. The application of the RSS was generally within the recommended range of malignancy, apart from the moderate and mild suspicion and not suspicious or benign classifications of ACR TI-RADS, the intermediate and low-risk classifications of EU TI-RADS, and the intermediate, low, and very low suspicion classifications of ATA TI-RADS. Beyond this, approximately 6.5% of the nodules (42/643) did not meet any of the classification standards in ATA TI-RADS, and showed a malignant risk of 57% (24 of 42 nodules). The other two RSSs (namely, BTA TI-RADS and FSE TI-RADS) did not propose malignancy risk estimates for each category (Table S3 [online]).

To test the efficiencies of the RSS diagnoses and evaluations, we compared the sensitivities, specificities, and positive and negative predictive values at an optimal diagnostic cut-off, as summarized in Table 3. The area under the receiver operating characteristic curve (AUC) of SMI TI-RADS (0.94 [95% CI: 0.92, 0.96]) was relatively enhanced, compared to that of C TI-RADS (0.86 [95% CI: 0.83, 0.89], p < 0.001), EU TI-RADS (0.84 [95% CI: 0.81, 0.86], p < 0.001), ACR TI-RADS (0.84 [95% CI: 0.81, 0.87], p < 0.001), KSThR TI-RADS (0.80 [95% CI: 0.77, 0.83], p < 0.001), AACE TI-RADS (0.86 [95% CI: 0.83, 0.88], p < 0.001), ATA TI-RADS (0.84 [95% CI: 0.81, 0.87], p < 0.001), FSE TI-RADS (0.82 [95% CI: 0.79, 0.85], p < 0.001), and BTA TI-RADS (0.76 [95% CI: 0.73, 0.80], p < 0.001) (Fig. S3 [online]). The SMI TI-RADS exhibited the largest malignancy biopsy yield at 79% (189/240 nodules), and was closely followed by ACR TI-RADS at 74% (190/258 nodules), AACE TI-RADS at 72% (197/273 nodules), FSE TI-RADS at 68% (191/280 nodules), EU TI-RADS at 66% (200/302 nodules), C TI-RADS at 64% (211/332 nodules), ATA TI-RADS at 61% (210/342 nodules), KSThR TI-RADS at 59% (211/355 nodules), and BTA TI-RADS at 59% (328/557 nodules) (Fig. S4 [online]).Table 3 Comparison of the diagnostic performances of TI-RADS

Guideline	Sensitivity, (%)	Specificity, (%)	PPV, (%)	NPV, (%)	AUC	p value*	
C TI-RADS	87 (84, 91) [301/345]	85 (81, 89) [252/298]	87 (83, 90) [301/347]	85 (81, 89) [252/296]	0.86 (0.83, 0.89)	< 0.001	
ACR TI-RADS	82 (78, 86) [284/345]	86 (82, 90) [256/298]	87 (84, 91) [284/326]	81 (76, 85) [256/317]	0.84 (0.81, 0.87)	< 0.001	
EU TI-RADS	89 (86, 93) [308/345]	78 (73, 83) [232/298]	82 (79, 86) [308/374]	86 (82, 90) [232/269]	0.84 (0.81, 0.86)	< 0.001	
KSThR TI-RADS	84 (80, 88) [289/345]	77 (72, 81) [228/298]	81 (76, 85) [289/359]	80 (76, 85) [228/284]	0.80 (0.77, 0.83)	< 0.001	
AACE TI-RADS	90 (87, 94) [312/345]	81 (76, 85) [241/298]	85 (81, 88) [312/369]	88 (84, 92) [241/274]	0.86 (0.83, 0.88)	< 0.001	
ATA TI-RADS	85 (81, 89) [273/321]	82 (78, 87) [230/280]	85 (81, 89) [273/323]	83 (78, 87) [230/278]	0.84 (0.81, 0.87)	< 0.001	
BTA TI-RADS	70 (65, 74) [240/345]	83 (79, 87) [247/298]	83 (78, 87) [240/291]	70 (65, 75) [247/352]	0.76 (0.73, 0.80)	< 0.001	
FSE TI-RADS	89 (85, 92) [306/345]	75 (70, 80) [224/298]	81 (77, 85) [306/380]	85 (81, 90) [224/263]	0.82 (0.79, 0.85)	< 0.001	
SMI TI-RADS	87 (84, 91) [301/345]	87 (83, 91) [259/298]	89 (85, 92) [301/340]	86 (82, 89) [259/303]	0.94 (0.92, 0.96)	NA	
The data in parentheses represent a 95% CI value; the data in brackets represent the corresponding number and total number of the nodule

C TI-RADS Chinese Society of Ultrasound in Medicine system, ACR TI-RADS the American College of Radiology Thyroid Imaging Reporting and Data System, EU TI-RADS European Thyroid Association Guidelines for Ultrasound Malignancy Risk Stratification of Thyroid Nodules, KSThR TI-RADS Korean Society of Thyroid Radiology Imaging Guidelines for Thyroid Nodules and Differentiated Thyroid Cancer, AACE TI-RADS American Association of Clinical Endocrinologists Guidelines for Clinical Practice for the Diagnosis and Management of Thyroid Nodules, ATA TI-RADS American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer, BTA TI-RADS British Thyroid Association 2014 classification ultrasound scoring Guidelines of thyroid nodules in predicting malignancy, FSE TI-RADS Guidelines of the French Society of Endocrinology for the management of thyroid nodules, PPV positive predictive value, NPV negative predictive value, AUC area under the curve, NA not applicable

* Compared the AUC between SMI TI-RADS and other TI-RADS systems

The SMI TI-RADS also displayed the least UBR at 21% (51/240 nodules), closely accompanied by ACR TI-RADS at 26% (68/258 nodules), AACE TI-RADS at 28% (76/273 nodules), FSE TI-RADS at 32% (89/280 nodules), EU TI-RADS at 34% (102/302 nodules), C TI-RADS at 36% (121/332 nodules), ATA TI-RADS at 39% (132/342 nodules), KSThR TI-RADS at 41% (144/355 nodules), and BTA TI-RADS at 41% (229/557 nodules). For the purposes of this study, nodules with indications for biopsy were considered positive and those without indications for biopsy were considered negative, their sensitivities, specificities, and positive and negative predictive values are presented in Table 4.Table 4 Comparison of the biopsy performance of TI-RADS

Guideline	Sensitivity, (%)	Specificity, (%)	PPV, (%)	NPV, (%)	AUC	BYM, (%)	UBR, (%)	
C TI-RADS	61 (56, 66) [211/345]	59 (54, 65) [177/298]	64 (58, 69) [211/332]	57 (51, 62) [177/311]	0.60 (0.56, 0.64)	64 (211/332)	36 (121/332)	
ACR TI-RADS	55 (50, 60) [190/345]	77 (72, 82) [230/298]	74 (68, 79) [190/258]	60 (55, 65) [230/385]	0.66 (0.63, 0.70)	74 (190/258)	26 (68/258)	
EU TI-RADS	58 (53, 63) [200/345]	66 (60, 71) [196/298]	66 (61, 72) [200/302]	58 (52, 63) [196/341]	0.62 (0.58, 0.66)	66 (200/302)	34 (102/302)	
KSThR TI-RADS	61 (56, 66) [211/345]	52 (46, 57) [154/298]	59 (54, 65) [211/355]	54 (48, 59) [154/288]	0.56 (0.53, 0.60)	59 (211/355)	41 (144/355)	
AACE TI-RADS	57 (52, 62) [197/345]	75 (70, 79) [222/298]	72 (67, 78) [197/273]	60 (55, 65) [222/370]	0.66 (0.62, 0.69)	72 (197/273)	28 (76/273)	
ATA TI-RADS	61 (56, 66) [210/345]	56 (50, 61) [166/298]	61 (56, 67) [210/342]	55 (50, 61) [166/301]	0.58 (0.54, 0.62)	61 (210/342)	39 (132/342)	
BTA TI-RADS	95 (93, 97) [328/345]	23 (18, 28) [69/298]	59 (55, 63) [328/557]	80 (72, 89) [69/86]	0.59 (0.56, 0.62)	59 (328/557)	41 (229/557)	
FSE TI-RADS	55 (50, 61) [191/345]	70 (65, 75) [209/298]	68 (63, 74) [191/280]	58 (53, 63) [209/363]	0.63 (0.59, 0.66)	68 (191/280)	32 (89/280)	
SMI TI-RADS	55 (50, 60) [189/345]	83 (79, 87) [247/298]	79 (74, 84) [189/240]	61 (57, 66) [247/403]	0.69 (0.65, 0.72)	79 (189/240)	21 (51/240)	
C TI-RADS Chinese Society of Ultrasound in Medicine system, ACR TI-RADS The American College of Radiology Thyroid Imaging Reporting and Data System, EU TI-RADS European Thyroid Association Guidelines for Ultrasound Malignancy Risk Stratification of Thyroid Nodules, KSThR TI-RADS Korean Society of Thyroid Radiology Imaging Guidelines for Thyroid Nodules and Differentiated Thyroid Cancer, AACE TI-RADS American Association of Clinical Endocrinologists Guidelines for Clinical Practice for the Diagnosis and Management of Thyroid Nodules, ATA TI-RADS American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer, BTA TI-RADS British Thyroid Association 2014 classification ultrasound scoring Guidelines of thyroid nodules in predicting malignancy, FSE TI-RADS Guidelines of the French Society of Endocrinology for the management of thyroid nodules, PPV positive predictive value, NPV negative predictive value, AUC area under the curve, BYM biopsy yield of malignancy, UBR unnecessary biopsy rate

Validation

In the internal validation cohort, the AUC, sensitivity, specificity, BYM, and UBR of the SMI TI-RADS were 0.89 (95% CI: 0.83, 0.94), 0.83 (95% CI: 0.76,0.91), 0.85 (95% CI: 0.75, 0.94), 79% (59/75 nodules), and 21% (16/75 nodules), respectively. In the external validation cohort, the AUC, sensitivity, specificity, BYM, and UBR of the SMI TI-RADS were 0.91 (95% CI: 0.85, 0.96), 0.89 (95% CI: 0.81, 0.97), 0.86 (95% CI: 0.76, 0.96), 72% (31/43 nodules), and 28% (12/43 nodules), respectively.

Evaluation of interobserver variability

Based on our interobserver variability assessment, the κ value was 0.63 for echogenicity, 0.72 for shape, 0.75 for margin, 0.52 for echogenic foci, 0.66 for extrathyroidal extension, 0.46 for vascularity at SMI mode, 0.55 for ring-SMI patterns, 0.70 for penetrating vascularity, and 0.60 for flow-signal enlarged at SMI, respectively (Table S4 [online]).

Discussion

To explore the significance of vascular characteristics in the stratification of malignancy risk in thyroid nodules, we redefined SMI features and combined them with conventional US features to establish the SMI TI-RADS. All features in our hierarchical system were qualitative features, and this point-based approach allowed all thyroid nodules to be classified, ensuring the practicality of this stratification system.

Malignant tumors are characterized by abnormal vascular distribution and the presence of irregular vasculature [26–30]. However, due to limitations in current imaging methods and use, they had not yet been included in the TI-RADS stratification system. In recent years, in response to the shortcomings of conventional US and CDFI imaging, imaging techniques capable of assessing low levels of blood flow and smaller blood vessels have been developed [31–34]. These have provided more detailed images of microvascular branches and the distribution of blood flow in both the nodule and the adjoining parenchyma, providing an accurate evidence-based foundation for the stratification of malignancy risk in thyroid nodules to better guide clinical diagnosis and treatment.

We thus constructed a novel system, using the addition of SMI features to predict thyroid nodule malignancy by the rounded regression coefficient, SMI TI-RADS, covering features such as SMI vascularity, ring-SMI patterns, penetrating vascularity, and flow-signal enlargement. We also provided a detailed explanation of the above features to increase the practicality of our system, and scored thyroid nodules directly to obtain the risk stratification. It was found that the AUC, sensitivity, and specificity of the SMI TI-RADS were 0.94 (95% CI: 0.92, 0.96), 0.87 (95% CI: 0.84, 0.91), and 0.87 (95% CI: 0.83, 0.91), respectively.

There is no clear consensus on the requirements for needle biopsy among the different TI-RADS stratification systems. Without biopsy, approximately 26.8‒27.7% of patients with thyroid nodules will need to undergo active surveillance for a long time, leading to a certain amount of psychological pressure in the patients [9]. Therefore, it is important to consider the size threshold for biopsy when developing TI-RADS risk stratification, seeking high specificity and minimizing unnecessary biopsies [35, 36]. Compared with the remaining eight RSSs, our results showed that the specificity, BYM, and UBR of SMI TI-RADS were 0.83 (95% CI: 0.79, 0.87), 79%, and 21%, respectively.

Our study has several limitations. First, the study was retrospective and selection bias is thus inevitable. As our institution is a Class A tertiary hospital, the majority of patients were likely referred from local centers or clinics after the discovery of suspicious findings. Furthermore, we only included patients with thyroid nodules who had received SMI- and US-guided FNA or surgical resection, which may have resulted in a higher likelihood of nodule malignancy. Second, owing to the principle of Doppler US imaging-based SMI examination, the blood flow display was predictably impacted by the nodule location depth and the cardiac cycle, which caused certain biases in the selection of data. Third, we did not include multicenter data. The accuracy of the SMI TI-RADS stratification system we constructed thus requires further verification using more imaging data, including multi-center training cohorts and validation cohorts. Fourth, the interpretation of all the indicators used for evaluation, such as conventional US, SMI features, and FNA results, was dependent on the doctor’s level of experience and may thus be subject to inter-observer differences.

In summary, we redefined SMI-related features and classification criteria, and successfully constructed the SMI TI-RADS stratification system based on ACR TI-RADS. We recommend additional prospective multicenter verification of our RSS prior to widespread clinical application.

Supplementary information

ELECTRONIC SUPPLEMENTARY MATERIAL

Abbreviations

AUC Area under the curve

BYM Biopsy yield of malignancy

FNA Fine-needle aspiration

MVUS Microvascular ultrasound

SMI Superb microvascular imaging

TI-RADS Thyroid imaging reporting and data system

UBR Unnecessary biopsy rate

US Ultrasound

Supplementary information

The online version contains supplementary material available at 10.1186/s13244-024-01806-5.

Acknowledgements

The authors thank Zexin Chen, PhD, for assistance in statistical analysis.

Authors contributions

Guarantors of the integrity of the entire study, G.M., L.C., T.Z., Y.H., and P.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of the final version of the submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, G.M., W.Y, Y.Z., X.W., Z.S., T.Z., and Y.H.; clinical studies, G.M., L.C., Z.L., Y.Z., X.W., T.Z., Y.H., and P.H.; statistical analysis, G.M., L.C., Y.W., Y.Z., Z.S., T.Z., and P.H.; and manuscript editing, G.M., L.C., Y.W., Y.Z., T.Z., P.H., and Y.H.

Funding

We greatly acknowledge the financial support from the National Natural Science Foundation of China (82030048, 82230069, 32201138, and 82302193), the Natural Science Foundation of Zhejiang Province (LQ24H180004 and LQ24H160010). The General Research Program of Health Commission in Zhejiang Province (2024KY1089 and 2024KY1502). The Ningbo Natural Science Foundation (20221JCGY010043).

Data availability

The datasets analyzed for the current study are not publicly available due to patient privacy. Reasonable requests will be reviewed individually for application to the corresponding author.

Declarations

Ethics approval and consent to participate

Institutional Review Board approval was obtained from the Second Affiliated Hospital of Zhejiang University School of Medicine (no. 2023-0944). Due to the retrospective nature of this study, the participant informed consent requirement was waived and this research received ethical approval from the Second Affiliated Hospital of Zhejiang University School of Medicine (no. 2023-0944).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Guangrong Ma and Libin Chen contributed equally to this work.
==== Refs
References

1. Chen DW Lang BHH McLeod DSA Newbold K Haymart MR Thyroid cancer Lancet 2023 401 1531 1544 10.1016/S0140-6736(23)00020-X 37023783
Chen DW, Lang BHH, McLeod DSA, Newbold K, Haymart MR (2023) Thyroid cancer. Lancet 401:1531–154437023783
2. Davies L Hoang JK Thyroid cancer in the USA: current trends and outstanding questions Lancet Diabetes Endocrinol 2021 9 11 12 10.1016/S2213-8587(20)30372-7 33220765
Davies L, Hoang JK (2021) Thyroid cancer in the USA: current trends and outstanding questions. Lancet Diabetes Endocrinol 9:11–1233220765
3. Ma C Zhang N Wang T Inflammatory cytokine-regulated LNCPTCTS suppresses thyroid cancer progression via enhancing snail nuclear export Cancer Lett 2023 575 216402 10.1016/j.canlet.2023.216402 37741431
Ma C, Zhang N, Wang T et al (2023) Inflammatory cytokine-regulated LNCPTCTS suppresses thyroid cancer progression via enhancing snail nuclear export. Cancer Lett 575:21640237741431
4. Haugen BR Alexander EK Bible KC 2015 American Thyroid Association Management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer Thyroid 2016 26 1 133 10.1089/thy.2015.0020 26462967
Haugen BR, Alexander EK, Bible KC et al (2016) 2015 American Thyroid Association Management guidelines for adult patients with thyroid nodules and differentiated thyroid cancer: the American Thyroid Association guidelines task force on thyroid nodules and differentiated thyroid cancer. Thyroid 26:1–13326462967
5. Zhou J Yin L Wei X 2020 Chinese guidelines for ultrasound malignancy risk stratification of thyroid nodules: the C-TIRADS Endocrine 2020 70 256 279 10.1007/s12020-020-02441-y 32827126
Zhou J, Yin L, Wei X et al (2020) 2020 Chinese guidelines for ultrasound malignancy risk stratification of thyroid nodules: the C-TIRADS. Endocrine 70:256–27932827126
6. Filetti S Durante C Hartl D Thyroid cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up† Ann Oncol 2019 30 1856 1883 10.1093/annonc/mdz400 31549998
Filetti S, Durante C, Hartl D et al (2019) Thyroid cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up†. Ann Oncol 30:1856–188331549998
7. Vaccarella S Franceschi S Bray F Wild CP Plummer M Dal Maso L Worldwide thyroid-cancer epidemic? The increasing impact of overdiagnosis N Engl J Med 2016 375 614 617 10.1056/NEJMp1604412 27532827
Vaccarella S, Franceschi S, Bray F, Wild CP, Plummer M, Dal Maso L (2016) Worldwide thyroid-cancer epidemic? The increasing impact of overdiagnosis. N Engl J Med 375:614–61727532827
8. Alexander EK Cibas ES Diagnosis of thyroid nodules Lancet Diabetes Endocrinol 2022 10 533 539 10.1016/S2213-8587(22)00101-2 35752200
Alexander EK, Cibas ES (2022) Diagnosis of thyroid nodules. Lancet Diabetes Endocrinol 10:533–53935752200
9. Ha SM Baek JH Na DG Diagnostic performance of practice guidelines for thyroid nodules: thyroid nodule size versus biopsy rates Radiology 2019 291 92 99 10.1148/radiol.2019181723 30777805
Ha SM, Baek JH, Na DG et al (2019) Diagnostic performance of practice guidelines for thyroid nodules: thyroid nodule size versus biopsy rates. Radiology 291:92–9930777805
10. Eszlinger M Lau L Ghaznavi S Molecular profiling of thyroid nodule fine-needle aspiration cytology Nat Rev Endocrinol 2017 13 415 424 10.1038/nrendo.2017.24 28361927
Eszlinger M, Lau L, Ghaznavi S et al (2017) Molecular profiling of thyroid nodule fine-needle aspiration cytology. Nat Rev Endocrinol 13:415–42428361927
11. Durante C Hegedüs L Na DG International expert consensus on US lexicon for thyroid nodules Radiology 2023 309 e231481 10.1148/radiol.231481 37906014
Durante C, Hegedüs L, Na DG et al (2023) International expert consensus on US lexicon for thyroid nodules. Radiology 309:e23148137906014
12. Ruan J Xu X Cai Y A practical CEUS thyroid reporting system for thyroid nodules Radiology 2022 305 149 159 10.1148/radiol.212319 35699576
Ruan J, Xu X, Cai Y et al (2022) A practical CEUS thyroid reporting system for thyroid nodules. Radiology 305:149–15935699576
13. Kobaly K Kim CS Mandel SJ Contemporary management of thyroid nodules Annu Rev Med 2022 73 517 528 10.1146/annurev-med-042220-015032 34416120
Kobaly K, Kim CS, Mandel SJ (2022) Contemporary management of thyroid nodules. Annu Rev Med 73:517–52834416120
14. Kim PH Yoon HM Baek JH Diagnostic performance of five adult-based US risk stratification systems in pediatric thyroid nodules Radiology 2022 305 190 198 10.1148/radiol.212762 35787203
Kim PH, Yoon HM, Baek JH et al (2022) Diagnostic performance of five adult-based US risk stratification systems in pediatric thyroid nodules. Radiology 305:190–19835787203
15. Trimboli P Castellana M Piccardo A The ultrasound risk stratification systems for thyroid nodule have been evaluated against papillary carcinoma. A meta-analysis Rev Endocr Metab Disord 2021 22 453 460 10.1007/s11154-020-09592-3 32959174
Trimboli P, Castellana M, Piccardo A et al (2021) The ultrasound risk stratification systems for thyroid nodule have been evaluated against papillary carcinoma. A meta-analysis. Rev Endocr Metab Disord 22:453–46032959174
16. De Nicola H Szejnfeld J Logullo AF Wolosker AM Souza LR Chiferi V Jr Flow pattern and vascular resistive index as predictors of malignancy risk in thyroid follicular neoplasms J Ultrasound Med 2005 24 897 904 10.7863/jum.2005.24.7.897 15972703
De Nicola H, Szejnfeld J, Logullo AF, Wolosker AM, Souza LR, Chiferi Jr V (2005) Flow pattern and vascular resistive index as predictors of malignancy risk in thyroid follicular neoplasms. J Ultrasound Med 24:897–90415972703
17. Cansu A Ayan E Kul S Eyüboğlu İ Oğuz Ş Mungan S Diagnostic value of 3D power Doppler ultrasound in the characterization of thyroid nodules Turk J Med Sci 2019 49 723 729 31203590
Cansu A, Ayan E, Kul S, Eyüboğlu İ, Oğuz Ş, Mungan S (2019) Diagnostic value of 3D power Doppler ultrasound in the characterization of thyroid nodules. Turk J Med Sci 49:723–72931203590
18. Wang XN Zhao Q Li DJ Quantitative evaluation of primary retinitis pigmentosa patients using colour Doppler flow imaging and optical coherence tomography angiography Acta Ophthalmol 2019 97 e993 e997 10.1111/aos.14047 30963731
Wang XN, Zhao Q, Li DJ et al (2019) Quantitative evaluation of primary retinitis pigmentosa patients using colour Doppler flow imaging and optical coherence tomography angiography. Acta Ophthalmol 97:e993–e99730963731
19. Xue N Li P Deng H Yi J Xie Y Zhang S The spoke wheel color Doppler blood flow signal is a specific sign of papillary thyroid carcinoma Front Endocrinol (Lausanne) 2022 13 1030143 10.3389/fendo.2022.1030143 36387883
Xue N, Li P, Deng H, Yi J, Xie Y, Zhang S (2022) The spoke wheel color Doppler blood flow signal is a specific sign of papillary thyroid carcinoma. Front Endocrinol (Lausanne) 13:103014336387883
20. Wang C Lin M Zhong L Tian C The value of TI-RADS combined with superb microvascular imaging in distinguishing thyroid nodules: a protocol for systematic review and meta-analysis PLoS One 2021 16 e0245035 10.1371/journal.pone.0245035 33411820
Wang C, Lin M, Zhong L, Tian C (2021) The value of TI-RADS combined with superb microvascular imaging in distinguishing thyroid nodules: a protocol for systematic review and meta-analysis. PLoS One 16:e024503533411820
21. Guo Y Wang X Wang L The value of superb microvascular imaging and contrast-enhanced ultrasound for the evaluation of neovascularization in carotid artery plaques Acad Radiol 2023 30 403 411 10.1016/j.acra.2022.08.001 36123231
Guo Y, Wang X, Wang L et al (2023) The value of superb microvascular imaging and contrast-enhanced ultrasound for the evaluation of neovascularization in carotid artery plaques. Acad Radiol 30:403–41136123231
22. Kurt SA Kayadibi Y Saracoglu MS Prediction of molecular subtypes using superb microvascular imaging and shear wave elastography in invasive breast carcinomas Acad Radiol 2023 30 14 21 10.1016/j.acra.2022.04.017 35659854
Kurt SA, Kayadibi Y, Saracoglu MS et al (2023) Prediction of molecular subtypes using superb microvascular imaging and shear wave elastography in invasive breast carcinomas. Acad Radiol 30:14–2135659854
23. Jiang L Zhang D Chen YN Yu XJ Pan MF Lian L The value of conventional ultrasound combined with superb microvascular imaging and color Doppler flow imaging in the diagnosis of thyroid malignant nodules: a systematic review and meta-analysis Front Endocrinol (Lausanne) 2023 14 1182259 10.3389/fendo.2023.1182259 37415660
Jiang L, Zhang D, Chen YN, Yu XJ, Pan MF, Lian L (2023) The value of conventional ultrasound combined with superb microvascular imaging and color Doppler flow imaging in the diagnosis of thyroid malignant nodules: a systematic review and meta-analysis. Front Endocrinol (Lausanne) 14:118225937415660
24. Cibas ES Ali SZ The 2017 Bethesda system for reporting thyroid cytopathology Thyroid 2017 27 1341 1346 10.1089/thy.2017.0500 29091573
Cibas ES, Ali SZ (2017) The 2017 Bethesda system for reporting thyroid cytopathology. Thyroid 27:1341–134629091573
25. Tessler FN Middleton WD Grant EG ACR thyroid imaging, reporting and data system (TI-RADS): white paper of the ACR TI-RADS Committee J Am Coll Radiol 2017 14 587 595 10.1016/j.jacr.2017.01.046 28372962
Tessler FN, Middleton WD, Grant EG et al (2017) ACR thyroid imaging, reporting and data system (TI-RADS): white paper of the ACR TI-RADS Committee. J Am Coll Radiol 14:587–59528372962
26. Wan P Chen F Liu C Kong W Zhang D Hierarchical temporal attention network for thyroid nodule recognition using dynamic CEUS imaging IEEE Trans Med Imaging 2021 40 1646 1660 10.1109/TMI.2021.3063421 33651687
Wan P, Chen F, Liu C, Kong W, Zhang D (2021) Hierarchical temporal attention network for thyroid nodule recognition using dynamic CEUS imaging. IEEE Trans Med Imaging 40:1646–166033651687
27. Carmeliet P Angiogenesis in health and disease Nat Med 2003 9 653 660 10.1038/nm0603-653 12778163
Carmeliet P (2003) Angiogenesis in health and disease. Nat Med 9:653–66012778163
28. Dudley AC Griffioen AW Pathological angiogenesis: mechanisms and therapeutic strategies Angiogenesis 2023 26 313 347 10.1007/s10456-023-09876-7 37060495
Dudley AC, Griffioen AW (2023) Pathological angiogenesis: mechanisms and therapeutic strategies. Angiogenesis 26:313–34737060495
29. Jiang Z Zhou J Li L Pericytes in the tumor microenvironment Cancer Lett 2023 556 216074 10.1016/j.canlet.2023.216074 36682706
Jiang Z, Zhou J, Li L et al (2023) Pericytes in the tumor microenvironment. Cancer Lett 556:21607436682706
30. Liao Q Shi H Yang J FTO elicits tumor neovascularization in cancer-associated fibroblasts through eliminating m(6)A modifications of multiple pro-angiogenic factors Cancer Lett 2024 592 216911 10.1016/j.canlet.2024.216911 38685450
Liao Q, Shi H, Yang J et al (2024) FTO elicits tumor neovascularization in cancer-associated fibroblasts through eliminating m(6)A modifications of multiple pro-angiogenic factors. Cancer Lett 592:21691138685450
31. Fu Z Zhang J Lu Y Clinical applications of superb microvascular imaging in the superficial tissues and organs: a systematic review Acad Radiol 2021 28 694 703 10.1016/j.acra.2020.03.032 32418782
Fu Z, Zhang J, Lu Y et al (2021) Clinical applications of superb microvascular imaging in the superficial tissues and organs: a systematic review. Acad Radiol 28:694–70332418782
32. Zhu C Zhong L Lin M Tian C Wang C The value of TI-RADS combined with superb micro-vascular imagine in distinguishing benign and malignant thyroid nodules: a meta-analysis PLoS One 2022 17 e0261521 10.1371/journal.pone.0261521 35041691
Zhu C, Zhong L, Lin M, Tian C, Wang C (2022) The value of TI-RADS combined with superb micro-vascular imagine in distinguishing benign and malignant thyroid nodules: a meta-analysis. PLoS One 17:e026152135041691
33. Trimboli P Castellana M Virili C Performance of contrast-enhanced ultrasound (CEUS) in assessing thyroid nodules: a systematic review and meta-analysis using histological standard of reference Radiol Med 2020 125 406 415 10.1007/s11547-019-01129-2 31970579
Trimboli P, Castellana M, Virili C et al (2020) Performance of contrast-enhanced ultrasound (CEUS) in assessing thyroid nodules: a systematic review and meta-analysis using histological standard of reference. Radiol Med 125:406–41531970579
34. Xiao F Li JM Han ZY Multimodality US versus thyroid imaging reporting and data system criteria in recommending fine-needle aspiration of thyroid nodules Radiology 2023 307 e221408 10.1148/radiol.221408 37367448
Xiao F, Li JM, Han ZY et al (2023) Multimodality US versus thyroid imaging reporting and data system criteria in recommending fine-needle aspiration of thyroid nodules. Radiology 307:e22140837367448
35. Durante C Grani G Lamartina L Filetti S Mandel SJ Cooper DS The diagnosis and management of thyroid nodules: a review JAMA 2018 319 914 924 10.1001/jama.2018.0898 29509871
Durante C, Grani G, Lamartina L, Filetti S, Mandel SJ, Cooper DS (2018) The diagnosis and management of thyroid nodules: a review. JAMA 319:914–92429509871
36. Singh Ospina N Iñiguez-Ariza NM Castro MR Thyroid nodules: diagnostic evaluation based on thyroid cancer risk assessment BMJ 2020 368 l6670 10.1136/bmj.l6670 31911452
Singh Ospina N, Iñiguez-Ariza NM, Castro MR (2020) Thyroid nodules: diagnostic evaluation based on thyroid cancer risk assessment. BMJ 368:l667031911452
