Relationship Between Morphologic Characteristics of Ultrasonic Calcification in Thyroid Nodules and Thyroid Carcinoma

A novel approach to quantify calcifications of thyroid nodules in US images based on deep learning: predicting the risk of cervical lymph node metastasis in papillary thyroid cancer patients

OBJECTIVE

Based on ultrasound (US) images, this study aimed to detect and quantify calcifications of thyroid nodules, which are regarded as one of the most important features in US diagnosis of thyroid cancer, and to further investigate the value of US calcifications in predicting the risk of lymph node metastasis (LNM) in papillary thyroid cancer (PTC).

METHODS

Based on the DeepLabv3+ networks, 2992 thyroid nodules in US images were used to train a model to detect thyroid nodules, of which 998 were used to train a model to detect and quantify calcifications. A total of 225 and 146 thyroid nodules obtained from two centers, respectively, were used to test the performance of these models. A logistic regression method was used to construct the predictive models for LNM in PTCs.

RESULTS

Calcifications detected by the network model and experienced radiologists had an agreement degree of above 90%. The novel quantitative parameters of US calcification defined in this study showed a significant difference between PTC patients with and without cervical LNM (p < 0.05). The calcification parameters were beneficial to predicting the LNM risk in PTC patients. The LNM prediction model using these calcification parameters combined with patient age and other US nodular features showed a higher specificity and accuracy than the calcification parameters alone.

CONCLUSIONS

Our models not only detect the calcifications automatically, but also have value in predicting cervical LNM risk of PTC patients, thereby making it possible to investigate the relationship between calcifications and highly invasive PTC in detail.

CLINICAL RELEVANCE STATEMENT

Due to the high association of US microcalcifications with thyroid cancers, our model will contribute to the differential diagnosis of thyroid nodules in daily practice.

KEY POINTS

• We developed an ML-based network model for automatically detecting and quantifying calcifications within thyroid nodules in US images. • Three novel parameters for quantifying US calcifications were defined and verified. • These US calcification parameters showed value in predicting the risk of cervical LNM in PTC patients.

Deep learning approaches for differentiating thyroid nodules with calcification: a two-center study

BACKGROUND

Calcification is a common phenomenon in both benign and malignant thyroid nodules. However, the clinical significance of calcification remains unclear. Therefore, we explored a more objective method for distinguishing between benign and malignant thyroid calcified nodules.

METHODS

This retrospective study, conducted at two centers, involved a total of 631 thyroid nodules, all of which were pathologically confirmed. Ultrasound image sets were employed for analysis. The primary evaluation index was the area under the receiver-operator characteristic curve (AUROC). We compared the diagnostic performance of deep learning (DL) methods with that of radiologists and determined whether DL could enhance the diagnostic capabilities of radiologists.

RESULTS

The Xception classification model exhibited the highest performance, achieving an AUROC of up to 0.970, followed by the DenseNet169 model, which attained an AUROC of up to 0.959. Notably, both DL models outperformed radiologists (P < 0.05). The success of the Xception model can be attributed to its incorporation of deep separable convolution, which effectively reduces the model's parameter count. This feature enables the model to capture features more effectively during the feature extraction process, resulting in superior performance, particularly when dealing with limited data.

CONCLUSIONS

This study conclusively demonstrated that DL outperformed radiologists in differentiating between benign and malignant calcified thyroid nodules. Additionally, the diagnostic capabilities of radiologists could be enhanced with the aid of DL.

Risk factor analysis and prediction model for papillary thyroid carcinoma with lymph node metastasis

Objective

We aimed to identify the clinical factors associated with lymph node metastasis (LNM) based on ultrasound characteristics and clinical data, and develop a nomogram for personalized clinical decision-making.

Methods

A retrospective analysis was performed on 252 patients with papillary thyroid carcinoma (PTC). The patient's information was subjected to univariate and multivariate logistic regression analyses to identify risk factors. A nomogram to predict LNM was established combining the risk factors. The performance of the nomogram was evaluated using receiver operating characteristic (ROC) curve, calibration curve, cross-validation, decision curve analysis (DCA), and clinical impact curve.

Results

There are significant differences between LNM and non-LNM groups in terms of age, sex, tumor size, hypoechoic halo around the nodule, thyroid capsule invasion, lymph node microcalcification, lymph node hyperechoic area, peak intensity of contrast (PI), and area under the curve (AUC) of the time intensity curve of contrast (P<0.05). Age, sex, thyroid capsule invasion, lymph node microcalcification were independent predictors of LNM and were used to establish the predictive nomogram. The ROC was 0.800, with excellent discrimination and calibration. The predictive accuracy of 0.757 and the Kappa value was 0.508. The calibration curve, DCA and calibration curve demonstrated that the prediction model had excellent net benefits and clinical practicability.

Conclusion

Age, sex, thyroid capsule invasion, and lymph node microcalcification were identified as significant risk factors for predicting LNM in patients with PTC. The visualized nomogram model may assist clinicians in predicting the likelihood of LNM in patients with PTC prior to surgery.

Leveraging deep learning to identify calcification and colloid in thyroid nodules

Background

Both calcification and colloid in thyroid nodules are reflected as echogenic foci in ultrasound images. However, calcification and colloid have significantly different probabilities of malignancy. We explored the performance of a deep learning (DL) model in distinguishing the echogenic foci of thyroid nodules as calcification or colloid.

Methods

We conducted a retrospective study using ultrasound image sets. The DL model was trained and tested on 30,388 images of 1127 nodules. All nodules were pathologically confirmed. The area under the receiver-operator characteristic curve (AUC) was employed as the primary evaluation index.

Results

The YoloV5 (You Only Look Once Version 5) transfer learning model for thyroid nodules based on DL detection showed that the average sensitivity, specificity, and accuracy of distinguishing echogenic foci in the test 1 group (n = 192) was 78.41%, 91.36%, and 77.81%, respectively. The average sensitivity, specificity, and accuracy of the three radiologists were 51.14%, 82.58%, and 61.29%, respectively. The average sensitivity, specificity, and accuracy of distinguishing small echogenic foci in the test 2 group (n = 58) was 70.17%, 77.14%, and 73.33%, respectively. Correspondingly, the average sensitivity, specificity, and accuracy of the radiologists were 57.69%, 63.29%, and 59.38%.

Conclusions

The study demonstrated that DL performed far better than radiologists in distinguishing echogenic foci of thyroid nodules as calcifications or colloid.

Association between macrocalcification and papillary thyroid carcinoma and corresponding valuable diagnostic tool: retrospective study

BACKGROUND

Microcalcifications are suggested to be an indicator of thyroid malignancy, especially for papillary thyroid carcinoma (PTC), nonetheless, the association between macrocalcification and PTC is underexplored. Furthermore, screening methods like ultrasonography and ultrasound-guided fine needle aspiration biopsy (US-FNAB) are limited in evaluating macro-calcified thyroid nodules. Thus, we aimed to investigate the relationship between macrocalcification and PTC. We also explored the diagnostic efficiency of US-FNAB and proto-Oncogene Proteins B-raf V600E (BRAF V600E) mutation in macro-calcified thyroid nodules evaluation.

METHODS

A retrospective research of 2645 thyroid nodules from 2078 participants was performed and divided into three groups as non-, micro-, and macro-calcified for further PTC incidence comparison. Besides, a total of 100 macro-calcified thyroid nodules with both results of US-FNAB and BRAF V600E mutation were screened out for subsequent evaluation of diagnostic efficiency.

RESULTS

Compared to non-calcification, macrocalcification showed a significantly higher incidence of PTC (31.5% vs. 23.2%, P<0.05). Additionally, when compared with a single US-FNAB, the combination of US-FNAB and BRAF V600E mutation showed better diagnostic efficiency in diagnosing macro-calcified thyroid nodule (area under the curve (AUC) 0.94 vs. 0.84, P=0.03), with a significantly higher sensitivity (100.0% vs. 67.2%, P<0.01) and a comparable standard of specificity (88.9% vs. 100.0%, P=0.13).

CONCLUSIONS

Occurrence of macrocalcification in thyroid nodules may suggest a high risk of PTC, and the combination of US-FNAB and BRAF V600E showed a greater value in identifying macro-calcified thyroid nodules, especially with significantly higher sensitivity.

TRIAL REGISTRATION

The Ethics Committee of The First Affiliated Hospital of Wenzhou Medical University (2018-026).

Explore the diagnostic performance of 2020 Chinese Thyroid Imaging Reporting and Data Systems by comparing with the 2017 ACR-TIRADS guidelines: a single-center study

OBJECTIVE

To compare the diagnostic efficacy of the Chinese Thyroid Imaging Reporting and Data Systems (C-TIRADS) with the well-accepted ACR-TIRADS guidelines in identifying benign from malignant thyroid nodules.

METHODS

A total of 2064 nodules were collected from 1627 patients undergoing thyroid ultrasonography in our center between October 2019 and November 2021. Nodules were divided into two groups: "≥1 cm" and "<1 cm". Ultrasound features of each nodule were observed and recorded by two physicians with more than 15 years of experience and classified according to the ACR-TIRADS and C-TIRADS guidelines, respectively.

RESULTS

The area under the curve of the ACR-TIRADS guideline was higher than that of the C-TIRADS guideline (0.922, P = 0.017), the specificity and positive predictive value of the C-TIRADS guideline were higher (81.64%, 88.72%, all P < 0.05), which was more significant in the subgroup of nodules <1 cm (P = 0.001). In addition, there was no statistical difference between the two guidelines in the diagnostic efficacy indicators for nodules ≥1 cm. The ACR-TIRADS effectively reduced unnecessary biopsies compared with the C-TIRADS (P < 0.05).

CONCLUSIONS

There was high agreement between the two guidelines for the diagnosis of thyroid nodules, C-TIRADS guidelines had a higher specificity and simplicity while were inferior to the ACR-TIRADS guidelines in terms of reducing the number of biopsies.

An Ultrasound-based Prediction Model for Occult Contralateral Papillary Thyroid Carcinoma in Adolescents and Young Adults

RATIONALE AND OBJECTIVES

To investigate the occult contralateral papillary thyroid carcinoma (PTC)-associated ultrasound (US) and clinical characteristics and establish a US-based model for the prediction of occult contralateral carcinoma in adolescents and young adults (AYAs) who were diagnosed with unilateral thyroid carcinoma preoperatively.

MATERIALS AND METHODS

From January 2015 to December 2020, patients who were diagnosed with unilateral thyroid carcinoma by preoperative US examination and underwent total thyroidectomy or thyroid lobectomy with more than 60 months of US follow-up at our hospital were retrospectively collected. Univariate and multivariate analyses were applied to identify the independent risk factors associated with occult contralateral PTC in AYAs, on which a prediction model was developed. The performance of the model was evaluated with accuracy, sensitivity, specificity, and the area under the receiver operating characteristic curve.

RESULTS

Occult contralateral PTC was found in 91 of 365 (24.9%) PTC patients with a median age at diagnosis of 26 years (interquartile range, 24-29 years). The multivariate analysis indicated that the presence of contralateral benign nodule, intra-tumoral calcification, and intraglandular dissemination were significantly associated with occult contralateral PTC in AYAs. The prediction model, which incorporated all independent predictors, yielded an area under the receiver operating characteristic curve of .661 (95% CI: .602-.719). The accuracy, sensitivity and specificity were 67.9%, 54.9%, and 72.3%, respectively.

CONCLUSION

The US-based prediction model proposed here exhibited a favorable performance for predicting occult contralateral PTC, which might be used to determine the appropriate extent of surgery for AYAs who had a preoperative diagnosis of unilateral thyroid carcinoma.

Role of echogenic foci in ultrasonographic risk stratification of thyroid nodules: Echogenic focus scoring in the American College of Radiology Thyroid Imaging Reporting and Data System

Background

Although echogenic foci may raise malignancy rates in thyroid nodules, the association between peripheral calcification or macrocalcification and thyroid carcinoma is controversial. We evaluated the malignancy probability of various echogenic foci and explored whether the method of determining a thyroid nodule’s point score in the echogenic focus category of the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS) is reasonable.

Methods

We retrospectively evaluated 819 patients with 852 nodules. The patterns of echogenic foci on ultrasonography were classified into the following four categories: punctate echogenic foci, macrocalcification, peripheral calcification, and multiple different types of echogenic foci. The core needle biopsy results were divided into two groups: benign and malignant or suspicious for malignancy.

Results

Among the 852 nodules, 471 (55.3%) had echogenic foci on ultrasonography. Of these nodules, there was no significant statistical difference in the malignant or suspicious for malignancy rate between nodules with peripheral calcification and those with macrocalcification [40.0% (8/20) vs. 30.6% (11/36), respectively; p = 0.474]. The incidence of malignancy or suspicious for malignancy for nodules with peripheral calcification, macrocalcification, or multiple different types of echogenic foci was significantly lower than the incidence for punctate echogenic foci alone, with odds ratios of 0.265 [95% confidence interval (CI): 0.105–0.667; p = 0.005], 0.175 (95% CI: 0.083–0.368; p = 0.000), and 0.256 (95% CI: 0.136–0.482; p = 0.000), respectively.

Conclusion

We found no significant statistical difference in the risk of malignancy or suspicious for malignancy rate between peripheral calcification and macrocalcification in thyroid nodules. We observed that nodules with multiple different types of echogenic foci were not associated with higher malignant or suspicious for malignancy rates compared with nodules with punctate echogenic foci alone.

Thyroid cancer in the Democratic Republic of the Congo: Frequency and risk factors

BACKGROUND

The prevalence of thyroid cancer is increasing steadily in most countries, partly due to better, earlier diagnosis. However, there is little data for developing countries, where the technical platform is often very limited, especially in Africa.

OBJECTIVES

To assess the frequency of thyroid cancer in the Democratic Republic of the Congo (DRC) and to analyze the epidemiological, clinical, and ultrasound risk factors.

MATERIAL AND METHODS

This is a multicenter cross-sectional study of 594 patients operated on for a thyroid mass from 2005 to 2019, in 35 centers in the DRC and for whom histopathological analyses were performed.

RESULTS

The frequency of thyroid cancers in our cohort was 20%, mostly in patients over the age of 40 (62% of patients). These cancers were mainly diagnosed at the clinical stage, due to the presence of palpable masses. Papillary cancer was the most common (67.2% of patients), followed by follicular cancer (28% of cases). We found a high prevalence of anaplastic cancer (7.6%). These frequencies are probably the consequence of the fact that histopathological analyses are not systematically performed in the DRC, but mostly on tissues that the thyroid surgeons suspect to be malignant. Age ≥60 years, the presence of adenopathies upon palpation or on ultrasound, the solid nature and hypoechogenicity of nodules, the presence of macronodules and calcifications were the factors independently associated with the diagnosis of cancer in the study population.

CONCLUSIONS

In this first study performed in the DRC, we have found that thyroid cancer is common. It is mainly detected at clinical stages, with patients over the age of 40 years and women being the most affected. The histopathology distribution differs from that in developed countries, with a lower prevalence of papillary cancer and a higher prevalence of the anaplastic type. In developing countries, it appears necessary to introduce the use of more precise diagnostic tools for thyroid cancer and also, to reinforce the improvement of known, controllable risk factors such as iodine deficiency.

Different sonographic features of peripheral thyroid nodule calcification and risk of malignancy: a prospective observational study

Purpose

To investigate the association of peripheral calcification, as well as its sonographic features, with thyroid nodule malignancy.

Material and methods

This study was prospectively conducted during 2015-2020 on patients diagnosed with thyroid nodule undergoing ultrasound-guided fine-needle aspiration in Shahid Beheshti teaching hospital or private offices in Babol, northern Iran. The ultrasonographic characteristics of the nodules, as well as the cytological findings, were recorded. Regression analysis was used to assess the relationship between sonographic results and malignancy. We also used receiver operator characteristics (ROC) analysis to estimate the ability of ultrasound to predict the characteristic features of malignancy, as estimated by the area under the curve (AUC).

Results

A total of 1857 thyroid nodules were finally included, of which 84 were peripherally calcified nodules. There was a significant positive association between the nodule malignancy and peripheral calcification (OR = 2.23, 95% CI: 1.13-4.35). In the nodules with peripheral calcification, significant positive associations were seen between malignancy and lobulated margin (OR = 3.85, 95% CI: 1.02-14.54) and solid composition (OR = 4.05, 95% CI: 0.99-16.53). The ROC analysis indicated that AUC for lobulated margin and solid composition was 63.8% and 66.5%, respectively, in predicting malignant thyroid nodules.

Conclusion

The findings showed that peripheral calcification on sonography can be a potential indicator of malignant thyroid nodules. Also, the presence of lobulated margin and/or solid composition, besides peripheral calcification, can be helpful in better distinguishing malignant from benign nodules.

Molecular Aspects of Thyroid Calcification

In thyroid cancer, calcification is mainly present in classical papillary thyroid carcinoma (PTC) and in medullary thyroid carcinoma (MTC), despite being described in benign lesions and in other subtypes of thyroid carcinomas. Thyroid calcifications are classified according to their diameter and location. At ultrasonography, microcalcifications appear as hyperechoic spots ≤ 1 mm in diameter and can be named as stromal calcification, bone formation, or psammoma bodies (PBs), whereas calcifications > 1 mm are macrocalcifications. The mechanism of their formation is still poorly understood. Microcalcifications are generally accepted as a reliable indicator of malignancy as they mostly represent PBs. In order to progress in terms of the understanding of the mechanisms behind calcification occurring in thyroid tumors in general, and in PTC in particular, we decided to use histopathology as the basis of the possible cellular and molecular mechanisms of calcification formation in thyroid cancer. We explored the involvement of molecules such as runt-related transcription factor-2 (Runx-2), osteonectin/secreted protein acidic and rich in cysteine (SPARC), alkaline phosphatase (ALP), bone sialoprotein (BSP), and osteopontin (OPN) in the formation of calcification. The present review offers a novel insight into the mechanisms underlying the development of calcification in thyroid cancer.

Is ultrasonographic evaluation sensitive enough to detect multicentric papillary thyroid carcinoma?

Background

This study aimed to investigate the preoperative ultrasonographic (US) evaluation of multicentric papillary thyroid carcinoma (PTC) and to evaluate the association of US findings with lymph node metastasis and extracapsular extension in PTC.

Methods

Preoperative US evaluations of patients with PTC who underwent total thyroidectomy were retrospectively investigated. Pathological perspectives and US features of PTC were analyzed. The sensitivity of US in detecting multicentric PTC was evaluated.

Results

The present study included 89 PTC patients who underwent total thyroidectomy. In total, 164 nodules were detected by preoperative US. Significant differences in US pattern were found between benign and malignant nodules. Of the 89 patients with PTC, 33 (37.08%) cases were confirmed as multicentric PTC by operation and pathological examination, 22 (66.67%) of which were bilateral. Before surgery, only 23 patients were suspected as multicentric PTC based on US findings. Pathological examination revealed that malignant nodules in 17 (51.51%) patients with multicentric PTC had been missed by preoperative US. The malignant nodules that went undetected by US were micronodulars (1-4 mm). Furthermore, ultrasonography was less sensitive for the diagnosis of metastatic lymph nodes in the neck. US had more than 80% sensitivity for detection of extracapsular extension of cases.

Conclusions

US evaluation is not sensitive enough to detect multicentric PTC. The minute size of some nodules in multicentric PTC, may lead to them being missed by US evaluation. Ultrasonography is an optional tool for the detection of extracapsular extension, but it is less sensitive for diagnosing lymph node metastasis.

Petal-Like Calcifications in Thyroid Nodules on Ultrasonography: A Rare Morphologic Characteristic of Calcification Associated With Aggressive Biological Behavior

This study investigated a rare ultrasonographically detected thyroid petal-like calcification and its relationship with thyroid carcinoma and biological behavior. We described the clinical and ultrasonographical features of thyroid nodules with petal-like calcifications in 18 patients undergoing thyroid surgery and cervical lymph node dissection. All of the thyroid nodules with petal-like calcifications were papillary thyroid carcinomas (PTCs). Of the 18 patients, 13 (72.2%) had cervical central lymph node metastasis, and five (27.8%) had cervical lateral lymph node metastasis. Petal-like calcifications occurred in malignant thyroid nodules with a high incidence of lymph node metastasis, which may be a specific ultrasonographic feature associated with the aggressive biological behavior of PTC.