Diagnostic Value of Sonographic Features in Distinguishing Malignant Partially Cystic Thyroid Nodules: A Systematic Review and Meta-Analysis

Comparative analysis of machine learning-based ultrasound radiomics in predicting malignancy of partially cystic thyroid nodules

OBJECTIVE

To investigate the application of machine learning (ML) model-based thyroid ultrasound radiomics in the evaluation of malignancy in partially cystic thyroid nodules (PCTNs).

METHODS

One hundred and ninety-two patients with 197 nodules PCTNs from January 2020 to December 2020 were retrospectively analyzed. Radiomics features were extracted based on hand-crafted features from the ultrasound images, and machine learning methods were used to build a classification model by radiomics features. The least absolute shrinkage and selection operator regression was applied to select the features of nonzero coefficients from radiomics features. The prediction performance of the established model was mainly evaluated by the area under the curve (AUC) and accuracy, sensitivity, and specificity.

RESULTS

Nineteen radiomics features were extracted from the original images for each nodule. Eight ML classifiers were able to differentiate malignancy in PCTNs. The AUC, accuracy, sensitivity, and specificity of k-Nearest Neighbor (KNN) model were 0.909, 82.95%, 83.33%, and 89.90%, respectively, on the test cohort. The comparative result showed statistically equivalent performance for thyroid nodule diagnosis based on image fusion and single image. In addition, the ML-Based ultrasound radiomics system showed a better AUC as compared with ACR TI-RADS model and the ultrasound features model.

CONCLUSION

The novel ultrasonic-based ML model has an important clinical value for predicting malignancy in PCTNs. It can provide clinicians with a preoperative non-invasive primary screening method for PCTN diagnosis to avoid unnecessary medical investment and improve treatment outcomes.

Ultrasound-Based Risk Stratification System for the Assessment of Partially Cystic Thyroid Nodules

OBJECTIVE

To develop and validate a risk stratification system for the prediction of malignancy in Partially Cystic Thyroid Nodules (PCTNs).

METHODS

From January 2020 to December 2021, we retrospectively reviewed the sonography data of PCTNs patients from Hangzhou TCM Hospital, Hangzhou First People's Hospital two medical centers. The independent risk factors of malignant PCTNs were evaluated using univariate and multivariate logistic regression analysis. The nomogram prediction efficiency was assessed using Area Under the Curve (AUC) and calibration curves. To determine the clinic value of the predictive model, Decision Curve Analysis (DCA) was used.

RESULTS

In this retrospective study, a total of 285 patients (301 PCTNs) were enrolled, 242 nodules were benign, and 59 nodules were malignant. Younger age, hypoechoic, irregular margin and microcalcifications were found to be the independent risk factors associated with malignant PCTNs. The AUC, sensitivity, and specificity were 0.860, 77.1%, and 84.7% in the training set and 0.897, 91.7%, and 87.0% in the external validation set, respectively. The total point of nomogram is greater than 161, which shows the best to predict the malignancy of PCTNs.

CONCLUSIONS

Our findings demonstrated that the risk stratification system for the assessment of PTCNs showed good prediction capacities.

SFE-AFCE-SFMN 2022 Consensus on the management of thyroid nodules: Initial work-up for thyroid nodules

The SFE-AFCE-SFMN 2022 consensus deals with the management of thyroid nodules, a condition that is a frequent reason for consultation in endocrinology. In more than 90% of cases, patients are euthyroid, with benign non-progressive nodules that do not warrant specific treatment. The clinician's objective is to detect malignant thyroid nodules at risk of recurrence and death, toxic nodules responsible for hyperthyroidism or compressive nodules warranting treatment. The diagnosis and treatment of thyroid nodules requires close collaboration between endocrinologists, nuclear medicine physicians and surgeons, but also involves other specialists. Therefore, this consensus statement was established jointly by 3 societies: the French Society of Endocrinology (SFE), French Association of Endocrine Surgery (AFCE) and French Society of Nuclear Medicine (SFMN); the various working groups included experts from other specialties (pathologists, radiologists, pediatricians, biologists, etc.). This section deals with the initial work-up for thyroid nodules in adult patients, including clinical and biological evaluation, standardized ultrasound characterization and EU-TIRADS-based nodule selection for fine-needle aspiration biopsy. Indications for thyroid core-biopsies or open surgical biopsies and for cross-sectional imaging of the neck and upper chest are also mentioned.

Approach to FNA of Thyroid Gland Cysts

Fine needle aspiration is a well-known procedure for the diagnosis and management of thyroid lesions, representing the first diagnostic tool for the definition of their nature. In clinical practice, a thyroid nodule can be classified as solid, cystic, and partially cystic based on its internal components. Different thyroid imaging reporting systems and cytologic diagnostic systems have focused their attention on solid nodules, which are more frequently linked with a malignant outcome. In fact, numerous papers demonstrated that nodules with microcalcifications, a taller-than-wide shape, hypoechogeneity, and irregular margins, are more likely to be malignant on histology. Nevertheless, according to the literature, the risk of malignancy in a partially cystic thyroid nodule ranges between 3.3 and 17-5%, including, for instance, the possible diagnosis of a cystic papillary thyroid carcinoma and other malignant entities. Therefore, in the current review article, we are going to discuss the approach to thyroid cystic lesions on fine needle aspiration cytology.

Construction and validation of BRAF mutation diagnostic model based on ultrasound examination and clinical features of patients with thyroid nodules

Introduction: Fine Needle Aspiration (FNA) is currently the most popular method for identifying benign and malignant thyroid nodules. However, its diagnostic sensitivity is sometimes limited, which makes it necessary to apply genetic testing and other modalities as a secondary diagnostic method. The diagnostic accuracy of thyroid nodule can be improved by combining mutations in the B-Raf proto-oncogene serine/threonine kinase (BRAF) with FNA. Thus, this study was conducted to create a nomogram diagnostic model based on the clinical and ultrasonic characteristics of patients with BRAF mutations to aid in the identification of benign and malignant thyroid nodules using FNA.

Methods: From April 2018 to December 2021, 275 patients with thyroid nodules who underwent ultrasonography and BRAF gene testing (137 positive and 138 negative) were included from Xianyang Central Hospital. The clinical and ultrasonic characteristics of the patients were used to develop a nomographic, diagnostic model of BRAF gene mutation, and to validate and evaluate the usefulness of the model.

Results: Independent risk factors for BRAF mutations included: focal strong echogenicity (microcalcifications, OR = 3.04, 95%CI = 1.41–6.58, p = 0.005), hypoechogenicity (OR = 3.8, 95%CI = 1.14–12.61, p = 0.029), lymph node metastases (OR = 3.54, 95%CI = 1.43–8.75, p = 0.006), margin (lobulated, OR = 3.7, 95%CI = 1.66–8.23, p = 0.001; extrathyroidal invasion, OR = 2.81, 95%CI = 1.11–7.06, p = 0.029), and shape (vertical position, OR = 2.7, 95%CI = 1.11–6.59, p = 0.029). The area under the curve (AUC) of the receiver operating characteristic (ROC) curve of the BRAF mutation diagnostic model constructed on these factors was 0.806 (95% CI = 0.754–0.851), and 39.5% was set as the threshold probability of making a clinical decision. The results of the validation and clinical utility evaluation showed that our model had good predictive performance and clinical application value.

Conclusion: Our nomogram diagnostic model based on clinical and ultrasound features of patients accurately predicted the possibility of BRAF gene mutations in patients with thyroid nodules.

Accuracy of Ultrasound Diagnosis of Thyroid Nodules Based on Artificial Intelligence-Assisted Diagnostic Technology: A Systematic Review and Meta-Analysis

BACKGROUND

Ultrasonography (US) is the most common method of identifying thyroid nodules, but US images require an experienced surgeon for identification. Many artificial intelligence (AI) techniques such as computer-aided diagnostic systems (CAD), deep learning (DL), and machine learning (ML) have been used to assist in the diagnosis of thyroid nodules, but whether AI techniques can improve the diagnostic accuracy of thyroid nodules still needs to be explored.

OBJECTIVE

To clarify the accuracy of AI-based thyroid nodule US images for differentiating benign and malignant thyroid nodules.

METHODS

A search strategy of "subject terms + key words" was used to search PubMed, Cochrane Library, Embase, Web of Science, China Biology Medicine (CBM), and China National Knowledge Infrastructure (CNKI) for studies on AI-assisted diagnosis of thyroid nodules based on US images. The summarized receiver operating characteristic (SROC) curve and the pooled sensitivity and specificity were used to assess the performance of the diagnostic tests. The quality assessment of diagnostics accuracy studies-2 (QUADAS-2) tool was used to assess the methodological quality of the included studies. The Review Manager 5.3 and Stata 15 were used to process the data. Subgroup analysis was based on the integrity of data collection.

RESULTS

A total of 25 studies with 17,429 US images of thyroid nodules were included. AI-assisted diagnostic techniques had better diagnostic efficacy in the diagnosis of benign and malignant thyroid nodules: sensitivity 0.88 (95% CI: (0.85-0.90)), specificity 0.81 (95% CI: 0.74-0.86), diagnostic odds ratio (DOR) 30 (95% CI: 19-46). The SROC curve indicated that the area under the curve (AUC) was 0.92 (95% CI: 0.89-0.94). Threshold effect analysis showed a Spearman correlation coefficient: 0.17 < 0.5, suggesting no threshold effect for the included studies. After a meta-regression analysis of 4 different subgroups, the results showed a statistically significant effect of mean age ≥50 years on heterogeneity. Compared with studies with an average age of ≥50 years, AI-assisted diagnostic techniques had higher diagnostic performance in studies with an average age of <50 years (0.89 (95% CI: 0.87-0.92) vs. 0.80 (95% CI: 0.73-0.88)), (0.83 (95% CI: 0.77-0.88) vs. 0.73 (95% CI: 0.60-0.87)).

CONCLUSIONS

AI-assisted diagnostic techniques had good diagnostic efficacy for thyroid nodules. For the diagnosis of <50 year olds, AI-assisted diagnostic technology was more effective in diagnosis.

Analysis of Differential Diagnosis of Benign and Malignant Partially Cystic Thyroid Nodules Based on Ultrasound Characterization With a TIRADS Grade-4a or Higher Nodules

Partially cystic thyroid nodules (PCTNs) are a kind of thyroid nodule with both solid and cystic components, and are usually misdiagnosed as benign nodules. The objective of this study was to determine the ultrasound (US) characterizations with a TIRADS Grade-4a or higher partially cystic thyroid nodules (PCTNs) which are associated with being malignant or benign. In this study, 133 PCTNs with a TIRADS Grade-4a or higher were enrolled; 83 were malignant and 50 were benign. TI-RADS classification can detect malignant PCTNs, and its sensitivity, specificity, positive predictive value, negative predictive value, and accuracy are 39.8%, 96.0%, 94.3%, 49.0%, and 60.9%, respectively. Univariate analyses revealed that nodule shape, margin, and structure were related to PCTNs' being benign and malignant, among which nodules taller-than-wide, with an irregular shape, non-smooth margin, eccentric sharp angle, or edge sharp angle were significantly associated with malignancy while ovoid to round nodules, smooth margins, multiple separation, and eccentric obtuse angle structures were significantly associated with a benign nature. For the solid part of PCTNs, its free margin, echo, and calcification are related to benign and malignant PCTNs. Among them, the free margin of the solid part is non-smooth, hypoechoic, and microcalcification, which are related to malignant PCTNs, while the free margin of the solid part is smooth, isoechoic, macrocalcification, non-calcification and are related to benign PCTNs. Calcification of solid part and free margin are important factors for predicting malignant PCTNs. In addition, nodules' composition, blood flow signal, and other factors had nothing to do with PCTNs' being benign or malignant. In the multivariate Logistic regression analysis, solid part calcification (OR: 17.28; 95%CI: 5.14~58.08) and free margin (OR: 3.18; 95%CI: 1.01~10.00) were revealed to be the strongest independent predictors for malignancy (P<0.05). Our study indicated that understanding the ultrasound characteristics of malignant PCTNs, to avoid misdiagnosed PCTNs patients, is important to make a precise diagnosis and prognosis of PCTNs.