Diagnosis Of Thyroid Nodules Research Articles

Contrast Enhancement Ultrasound Improves Diagnostic Accuracy for Thyroid Nodules: A Prospective Multicenter Study

Abstract Objective To evaluate potential improvements in the diagnosis of thyroid nodules when conventional ultrasound (US) is combined with contrast-enhanced US (CEUS). Methods We recruited 515 participants with 323 malignant and 192 benign nodules, who underwent both US and CEUS examinations at 8 different medical centers in China between October 2020 and October 2021. We assessed the malignancy of thyroid nodules in US using the American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TIRADS). Diagnostic criteria for US and US + CEUS were developed by investigators based on evaluations of sonographic features. Using multivariate logistic regression and receiver operating characteristic (ROC) analysis, we compared diagnostic performance between the 2 methods based on criteria identified by investigators and via statistical models. Results On the basis of diagnostic criteria identified by investigators, we measured statistically significant differences in area under the curve (AUC) values between ACR TIRADS (0.83) and CEUS TIRADS (0.87; P < .001). On the basis of diagnostic regression models, we found statistically significant differences in AUC values between US (0.76) and US + CEUS (0.84; P = .001). Models based on US + CEUS outperformed those based on US alone (Akaike information criterion of 347.7 and significant improvement in integrated discrimination). These results were confirmed by similar analyses applied to a validation cohort. Conclusion The accuracy of conventional US for differentiating between benign and malignant thyroid nodules can be improved by combining this approach with CEUS.

A knowledge-interpretable multi-task learning framework for automated thyroid nodule diagnosis in ultrasound videos

Ultrasound has become the most widely used modality for thyroid nodule diagnosis, due to its portability, real-time feedback, lack of toxicity, and low cost. Recently, the computer-aided diagnosis (CAD) of thyroid nodules has attracted significant attention. However, most existing techniques can only be applied to either static images with prominent features (manually selected from scanning videos) or rely on ’black boxes’ that cannot provide interpretable results. In this study, we develop a user-friendly framework for the automated diagnosis of thyroid nodules in ultrasound videos, by simulating the typical diagnostic workflow used by radiologists. This process consists of two orderly part-to-whole tasks. The first interprets the characteristics of each image using prior knowledge, to obtain corresponding frame-wise TI-RADS scores. Associated embedded representations not only provide diagnostic information for radiologists but also reduce computational costs. The second task models temporal contextual information in an embedding vector sequence and selectively enhances important information to distinguish benign and malignant thyroid nodules, thereby improving the efficiency and generalizability of the proposed framework. Experimental results demonstrated this approach outperformed other state-of-the-art video classification methods. In addition to assisting radiologists in understanding model predictions, these CAD results could further ease diagnostic workloads and improve patient care.

Stratifying Indeterminate Cytology Thyroid Nodules by Combining Thyroid Imaging Reporting and Data Systems (TI-RADS) and The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC)

Objective: To determine the risk of malignancy of Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) indeterminate Thyroid Nodules (Bethesda III, IV and V) by combining cytologic (TSBRTC) and Thyroid Imaging Reporting and Data Systems (TI-RADS) ultrasonographic features based on final histopathology. Methods:Design: Retrospective review of recordsSetting: Tertiary Private Training HospitalParticipants: 551 recordsResults: Among 81 eligible participants, 59 out of 84 nodules (70.24%) wer malignant on histopathology. The malignancy risk of Bethesda classification was 60.87% (28 out of 46) for Bethesda III, 57.14% (8 out of 14) for Bethesda IV and 95.83% for Bethesda V. The malignancy risk for TI-RADS categories was 0 % (0/1) for TI-RADS 2, 50% (10 out of 20) for TI-RADS 3, 71.05 % for TI-RADS 4 and 91.67 % for TI-RADS 5. The highest risk of malignancy (100%) was associated with [Bethesda IV/TI-RADS 1, 2, and 3], [Bethesda V/TI-RADS 1, 2 and 3 [Bethesda IV and V/TI-RADS 1, 2 and 3] and [Bethesda IV/TI-RADS 5]. The lowest risk of malignancy (33.33%) was associated with [Bethesda III/TI-RADS1, 2 and 3]. A high Bethesda classification (Bethesda V) was almost 5x more likely to have a malignant anatomorphology compared with Bethesda III (p = .05) while a TI-RADS 4 or 5 category was almost 5x more likely to have a malignant anatomorphology compared to TI-RADS 1, 2 or 3 (p = .026). Conclusion: This study showed that TI-RADS scoring is a sensitive diagnostic classification in recognizing patients with thyroid cancer and combining Bethesda classification and TI-RADS scoring increases the sensitivity in the diagnosis of malignant thyroid nodules. A higher likelihood of malignancy is associated with higher Bethesda classification and TI-RADS scoring.

DIAGNOSTIC ACCURACY OF ULTRASONOGRAPHY IN DIFFERENTIATING BENIGN AND MALIGNANT THYROID NODULES

This study aimed to compare the effectiveness of ultrasonography and fine-needle aspiration cytology (FNAC) in distinguishing between benign and malignant thyroid nodules. To conduct this cross-sectional study, we obtained ethical approval from the institutional review board and collected data from KRL Hospital between December 1, 2021, and April 30, 2022. We included patients of either gender, aged 18 years or above, referred to the radiology department for diagnosis of benign or malignant thyroid nodules or FNAC of nodules, using non-probability consecutive sampling. After giving informed consent, all patients underwent high-resolution Gray ultrasonography performed by a consultant radiologist. The results showed that ultrasonography had a sensitivity of 90.62%, specificity of 88.37%, positive predictive value of 85.29%, negative predictive value of 92.68%, and an overall accuracy of 89.33% in detecting benign and malignant thyroid nodules, as compared to the gold standard, FNAC. Based on these findings, we conclude that ultrasonography is an excellent non-invasive diagnostic tool, with high accuracy and no invasiveness, making it the preferred method for identifying malignant thyroid nodules.

A Systematic Review and Meta-analysis on Ultrasound Detection of Thyroid Cancer in China

Background: One major drawback of using ultrasound for diagnosing thyroid nodules is its limited ability to distinguish between benign and malignant nodules. In China, the common methods for risk stratification and guiding fine needle aspiration (FNA) in diagnosing thyroid nodules are the Chinese Thyroid Imaging Reports and Data Systems (C-TIRADS) and American College of Radiology-Thyroid Imaging Reporting and Data System (ACR-TIRADS). Objectives: This review seeks to assess the effectiveness of C-TIRADS and ACR-TIRADS in accurately identifying the risk of malignancy in Chinese patients suspected of thyroid cancer. Methods: A detailed search was conducted in PubMed, Google Scholar, Medline, Embase, Web of Science, Cochrane, and China National Knowledge Infrastructure (CNKI) databases from January 2018 to December 2022. The analysis only considered original articles from China reporting the use of C-TIRADS and ACR-TIRADS confirmed by histology and FNA. Results: This review analyzed 26 studies with a total of 23,064 thyroid nodules from 19,114 patients to compare the diagnostic performance of C-TIRADS and ACR-TIRADS in predicting malignancy risk in thyroid nodules. Although the malignancy rates in each risk category were similar between the two systems, the TIRADS showed better diagnostic performance than C-TIRADS in terms of pooled specificity (95.0 % vs. 66.8 % of C-TIRADS). However, the pooled analysis showed that C-TIRADS had a better pooled sensitivity (94.6 % vs. 76.5% of ACR-TIRADS). The diagnostic odds ratio was 1.37 (95 % CI: 0.75-2.51) for ACR-TIRADS and 0.89 (95 % CI: 0.36-2.16) for C-TIRADS. Conclusion: Based on the results, both C-TIRADS and ACR-TIRADS are effective in predicting the risk of malignancy in thyroid nodules with similar overall diagnostic accuracy. The combination of both systems can be beneficial in enhancing accuracy in suspicious or uncertain cases. The long-term experience of the trained radiologists can readily help in concluding the diagnosis. As no single system or combination of systems can provide a 100% accurate prediction of the malignancy of thyroid nodules, the ultimate diagnosis relies on the concluding assessment of experienced radiologists and the medical team.

Ultrasound Image Classification of Thyroid Nodules Based on Attention Mechanism

Ultrasound imaging of the thyroid gland is currently the most common diagnostic technique, used for pre diagnosis of benign and malignant thyroid nodules. During the clinical diagnosis, the condition judgment requires the high clinical experience of physicians. The low level of medical care in remote areas has led to an increased rate of misdiagnosis of benign and malignant nodules. To improve the accuracy of diagnosis and treatment effectiveness of doctors, this work proposes a benign and malignant thyroid classification model based on mechanisms of attention. First, this paper performs data cleaning on the collected data and performs pre-processing operations such as data enhancement to construct a new dataset. Then, a network model based on Lightweight Global Attention Module (LGAM) is constructed for the benign and malignant classification of nodules. Finally, it is shown by experimental analysis and comparison that the improved LGAM-based thyroid node classification model has achieved better experimental results on the dataset and can achieve the optimal classification accuracy of 95.16% for the network model proposed in this work.

Monitoring Thyrotropin in Veterans With Thyroid Nodules.

After the initial thyroid nodule diagnosis, a patient's thyrotropin is often monitored. However, the American Thyroid Association guidelines do not offer recommendations for follow-up thyrotropin testing for patients with thyroid nodules who have no history of conditions or known medications that affect thyroid hormone levels. At the Veterans Affairs Dayton Healthcare System in Ohio, we conducted a retrospective chart review from January 2010 to December 2016 of 100 patients diagnosed with ≥ 1 thyroid nodule on imaging studies who had normal blood thyrotropin at the time of nodule diagnosis. The thyrotropin value was studied at and after diagnosis. A 95% CI was determined for the true population rate of patients with an abnormal thyrotropin at their most recent testing. χ2 tests for categorical variables and independent sample t tests for continuous variables were used to compare the abnormal and normal most recent thyrotropin groups. One hundred patients (male [83%], White race [82%]) with normal thyrotropin at nodule diagnosis had thyrotropin monitoring for a mean (SD) of 5.7 (2.5) years. Six of 100 patients (6%; 95% CI, 2.5%-12.7%) developed abnormal thyrotropin levels in a mean (SD) of 6.9 (3.1) years. When comparing the 6 patients with abnormal thyrotropin vs the 94 with normal thyrotropin, there were no significant differences in sex (P = .99), race (P = .55), age at diagnosis (P = .12), initial thyrotropin level (P = .24), most recent thyrotropin level (P = .98), or time from diagnosis to most recent thyrotropin level (P = .23). This study found no significant change in thyrotropin levels over time in patients with thyroid nodules and no history of medical conditions or medications known to affect thyrotropin levels. Monitoring thyrotropin over time may not be required in these patients. More studies are needed to provide additional data on thyrotropin monitoring for thyroid nodules so that clinicians can make evidence-based decisions.

Automatic ultrasound diagnosis of thyroid nodules: a combination of deep learning and KWAK TI-RADS

Objective. There has been a considerable amount of computer-aided diagnosis (CAD) methods highlighted in the field of ultrasonic examination (USE) of thyroid nodules. However, few researches focused on the automatic risk classification, which was the basis for determining whether fine needle aspiration (FNA) was needed. The aim of this work was to implement automatic risk level assessment of thyroid nodules. Approach. Firstly, 1862 cases of thyroid nodules with the results of USE and FNA were collected as the dataset. Then, an improved U-Net++ model was utilized for segmenting thyroid nodules in ultrasound images automatically. Finally, the segmentation result was imported into a multi-task convolutional neural network (MT-CNN), the design of which was based on the clinical guideline called KWAK TI-RADS. Apart from the category of benign and malignant, the MT-CNN also exported the classification result of four malignant features, solid component (SC), hypoechogenicity or marked hypoechogenicity (HMH), microlobulated or irregular margin (MIM), microcalcification (MC), which were used for counting the risk level in KWAK TI-RADS. Main results. The performance of the improved U-Net++ was evaluated on our test set, including 302 cases. The Dice coefficient and intersection over union reached 0.899, 0.816, respectively. The classification accuracy rates of SC, HMH, MIM, MC, were 94.5%, 92.8%, 86.1%, 88.9%, while the false positive (FP) rate was 6.0%, 5.6%, 10.6%, 12.9% respectively. As for the category of benign and malignant, the precision and recall rates were 93.7% and 94.4%. Significance. The proposed CAD method showed favourable performance in the diagnosis of thyroid nodules. Compared with other methods, it could provide reports closer to clinical practice for doctors.

Comparison of six machine learning methods for differentiating benign and malignant thyroid nodules using ultrasonographic characteristics

BackgroundSeveral machine learning (ML) classifiers for thyroid nodule diagnosis have been compared in terms of their accuracy, sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV), and area under the receiver operating curve (AUC). A total of 525 patients with thyroid nodules (malignant, n = 228; benign, n = 297) underwent conventional ultrasonography, strain elastography, and contrast-enhanced ultrasound. Six algorithms were compared: support vector machine (SVM), linear discriminant analysis (LDA), random forest (RF), logistic regression (LG), GlmNet, and K-nearest neighbors (K-NN). The diagnostic performances of the 13 suspicious sonographic features for discriminating benign and malignant thyroid nodules were assessed using different ML algorithms. To compare these algorithms, a 10-fold cross-validation paired t-test was applied to the algorithm performance differences.ResultsThe logistic regression algorithm had better diagnostic performance than the other ML algorithms. However, it was only slightly higher than those of GlmNet, LDA, and RF. The accuracy, sensitivity, specificity, NPV, PPV, and AUC obtained by running logistic regression were 86.48%, 83.33%, 88.89%, 87.42%, 85.20%, and 92.84%, respectively.ConclusionsThe experimental results indicate that GlmNet, SVM, LDA, LG, K-NN, and RF exhibit slight differences in classification performance.

BRAFV600E genetic testing should be recommended for Bethesda III or V thyroid nodules based on fine-needle aspiration

The preoperative diagnosis of thyroid nodules now routinely includes BRAFV600E genetic testing in most provincial and municipal hospitals in China. This study identified the most suitable patients of thyroid nodule for BRAFV600E genetic testing. We retrospectively collected data of patients from the Hospital Information System that had undergone fine needle aspiration biopsy (FNAB) from May 2019 to December 2021. Data of FNAB, BRAFV600E genetic testing, and post-surgical pathological diagnosis were collected. A total of 12,392 patients were included in this study. Among them, 7,010 patients underwent solely FNAB, while 5,382 patients had both FNAB and BRAFV600E genetic testing. In the FNAB group, 2,065 thyroid nodules were surgically removed, with a 93.12% malignancy rate. In the FNAB + BRAF group, 2,005 nodules were dissected, and the malignancy rate was 98.20%. However, it was evident that in the subgroups, the combination of FNAB and BRAFV600E genetic testing only benefited Bethesda III (p < 0.001) and V (p = 0.001) nodules. Overall, the combination of FNAB with BRAFV600E genetic testing significantly improved the malignancy rate of surgical thyroid nodes at our hospital when compared to FNAB alone. The subgroup analysis showed that BRAFV600E genetic testing only benefited Bethesda III and V nodules. These findings provide a clinical reference for rationally selecting the most suitable population for BRAFV600E genetic testing.

THU676 Real-time Assessment Of The Beneficial Role Of Computer-Aided Diagnosis In The Diagnosis Of Thyroid Nodules On Ultrasound

Abstract Disclosure: Y. Kim: None. S. Shin: None. E. Lee: None. D. Kim: None. J. Kwak: None. Objective: Computer-aided Diagnosis (CAD) has been applied in various medical fields. SERA (SEveRance Artificial intelligence program) is a deep convolutional neural network algorithm. We conducted this study to evaluate whether CAD-SERA helps diagnose the malignancy when the ultrasound was used by physicians with low levels of experience and to compare with the experienced radiologist. Methods: In real-time, an inexperienced physician and a radiologist simultaneously assessed the same thyroid nodules to determine malignancy and performed CAD. The diagnosis was changed if it was necessary to correct it by referring to the CAD results, and this was compared with the surgical pathology or biopsy pathology. Results: We evaluated 191 thyroid nodules in 179 patients. Area Under the Receiver Operating Characteristics (AUROC) was analyzed with the diagnoses with/without CAD assistance and pathology results. The AUROC of the inexperienced physician was 0.716, and when assisted by CAD that increased to 0.740 (Accuracy 73.82 -&amp;gt; 76.44%). In the case of the radiologist, AUROC without CAD assist was 0.868 and 0.880 with the assist of CAD (Accuracy 84.29 -&amp;gt; 85.86%). Conclusion: CAD improves diagnostic performance in the diagnosis of thyroid nodules in various experienced physicians in ultrasound. In particular, the extent of this is greater in physicians with less experience in ultrasound. Presentation: Thursday, June 15, 2023

OR32-02 Thyroid Cancer Polygenic Risk Score Improves Risk Stratification Of Thyroid Nodules When Added To Ultrasound Imaging

Abstract Disclosure: N. Pozdeyev: None. M. Dighe: None. M. Barrio: None. C. Raeburn: None. H.A. Smith: None. M. Fisher: None. S. Chavan: None. N. Rafaels: None. J. Shortt: None. M. Leu: None. T. Clark: None. C. Marshall: None. B.R. Haugen: None. D. Subramanian: None. K. Crooks: None. C. Gignoux: None. T.A. Cohen: None. Purpose. Evaluating thyroid nodules to rule out malignancy is a common clinical task. Image-based risk stratification schemas rely on the presence of high-risk thyroid nodule sonographic features and, therefore, are less suitable for the diagnosis of malignant thyroid nodules that have a benign appearance on the ultrasound. To mitigate the deficiency of thyroid nodule evaluation relying solely on the sonographic characteristics, we used thyroid cancer polygenic risk score (PRS) to complement deep learning analysis of ultrasound images. Methods. A supervised deep learning classifier of thyroid nodules was trained on 32,545 thyroid US images from 621 nodules and tested on an independent set of 232 nodules from patients genotyped on the Illumina's MEGAEX platform. The deep-learning thyroid nodule classifier was developed by fine-tuning a BiT-M ResNet-50x1 convolutional neural network (CNN) pre-trained on the ImageNet-21k dataset. A polygenic risk score (PRS) was calculated using thyroid cancer genome-wide association meta-analysis summary statistics from the Global Biobank Meta-analysis Initiative. The thyroid cancer PRS was defined as a weighted sum of five alleles with the strongest association with thyroid cancer. CNN predictions and PRS were combined into a meta-classifier using logistic regression with or without genetic ancestry and demographic covariates. Results. The CNN classifier achieved an area under the receiver operating characteristic curve (AUC) of 0.83 on the out-of-sample test set of 232 thyroid nodules. The CNN classifier incorrectly classified thyroid nodules without suspicious sonographic characteristics belonging to difficult-to-diagnose subtypes such as follicular thyroid cancer and follicular variant of papillary thyroid cancer. Combining predictions from the CNN classifier with PRS into a cross-validated classifier improved AUC to 0.868 (DeLong test, p = 0.05). Incorporating genetic ancestry in the form of five genetic principal components further improved AUC of the benign vs. malignant CNN + PRS thyroid nodule classifier to 0.885 (p = 0.007). Finally, when age, sex, and nodule dimensions were considered, the AUC of the meta-classifier increased to 0.915 (p = 2.3e-4). The meta-classifier including predictions from CNN, PRS, and genetic principal components showed a sensitivity of 0.95, specificity of 0.61, NPV of 0.97, and PPV of 0.5. This performance was superior to that of the clinical Thyroid Imaging Reporting and Data System (TI-RADS) as reported by radiologists in ultrasound reports. Conclusions. For the first time, we showed PRS provides an orthogonal thyroid cancer risk assessment complementary to the ultrasound image-based thyroid nodule risk evaluation. This proof-of-concept study opens an opportunity for developing next-generation schemas for thyroid nodule evaluation incorporating clinical, imaging, and genetic data. Presentation: Sunday, June 18, 2023

FRI368 Thyroid Paraganglioma: A Rare Manifestation Of Paraganglioma Syndrome Associated With SDHD Mutation

Abstract Disclosure: K. Samuel: None. C. McMullen: None. C. Veloski: None. S. Kushchayev: None. J.E. Hallanger-Johnson: None. J. Hernandez Prera: None. J. Muzaffar: None. G.Q. Yang: None. V. Tarasova: None. Background: Thyroid paragangliomas (TP) are exceptionally rare, representing 0.5% of all paragangliomas (PGL) with total of less than 70 cases reported to date. We report a case of a TP presenting with an indeterminate thyroid nodule, as a first manifestation of multifocal head and neck PGL, associated with succinate dehydrogenase complex subunit D (SDHD) mutation. Clinical case: A 59-year female was initially seen at the tertiary referral center for an indeterminate thyroid nodule discovered incidentally on neck ultrasound (US) ordered for evaluation of a palpable posterior left neck mass. Neck/thyroid US showed multiple nonspecific bilateral lymph nodes measuring up to 1.2 cm and a 1.7 cm right interpolar markedly hypoechoic and hypervascular solid thyroid nodule (TR5). Clinically, the patient did not have symptoms related to the thyroid nodules or lymph nodes in the neck. Thyroid function tests were within normal range. An US-guided FNA biopsy was nondiagnostic. Core biopsy was suspicious for papillary thyroid carcinoma (hypercellular thyroid parenchyma with some follicular cells with nuclear irregularity, grooves, and hyperchromasia). Pathology slides review at the tertiary center showed nondiagnostic sample (Bethesda I) on cytopathology and neuroendocrine neoplasm consistent with PGL on core biopsy (immunohistochemical stains showed tumor cells positive for synaptophysin, chromogranin and GATA3. TTF1, TG, calcitonin, CEA and AE1/3 negative). US of the neck/thyroid 4 months later showed stable disease. Calcitonin and 24-h urinary catecholamines, metanephrines were within normal limits. There was no family history of pheochromocytoma or PGL. Genetic testing showed germline SDHD mutation (p.P81L(c.242c&amp;gt;5). 68Ga-DOTATATE PET/CT revealed multifocal areas of increased somatostatin receptor expression from the skull base to the left thoracic inlet, which included right and left skull base lesions, left neck level IIA lymph node, right and left level III lymph nodes, right thyroid lobe lesion, and left prevascular superior mediastinal lesion. An MRI of the brain and neck with and without contrast showed 1.7 cm right jugular, 0.6 cm right carotid, and 0.6 cm left vagal PGL, 1.3 cm right thyroid nodule, left level II 1.1x.6 cm lymphadenopathy, and 0.4 cm left superior mediastinal nodule. A CT of the neck w/contrast showed multiple areas likely representing bilateral carotid body tumors, left vagal PGL, right glomus jugular, and thyroid PGL stable over 6 months. MIBG nuclear medicine scan revealed no radiotracer accumulation in the lesions. EBRT therapy is planned to the skull base lesion. Conclusions: TP should be considered in the differential diagnosis of hypervascular thyroid nodules in patients with SDH-related pheochromocytoma-PGL syndromes. This case illustrates the importance of experienced multidisciplinary team evaluation and management of complex and rare PRG cases. Presentation: Friday, June 16, 2023

SAT467 Monitoring TSH Over Time May Not Be Necessary in Patients With Thyroid Nodules

Abstract Disclosure: S. Kaul: None. A. Gupta: None. Introduction: A substantial number of patients with thyroid nodules receive monitoring of blood thyroid stimulating hormone (TSH) levels after the initial diagnosis. There is not enough literature supporting long-term monitoring of blood TSH levels in these patients with thyroid nodules. The objective of this study was to evaluate the change in blood TSH level in patients with known thyroid nodules over time and to determine if repeat blood TSH testing is required in these patients. We hypothesized that monitoring of blood TSH level is not warranted in patients with thyroid nodules in the absence of a history of medical conditions or medications that are known to affect thyroid hormone levels. Methods: We studied 100 patients aged 18 to 99 years at the Dayton VA Medical Center with at least one thyroid nodule on imaging studies from January 2010 to December 2016 with normal blood TSH level at time of the diagnosis of the thyroid nodule. Subjects with abnormal thyroid hormone level, history of hot thyroid nodule, autoimmune thyroid disease (on imaging or blood work), thyroid or pituitary surgery, radioactive iodine treatment, neck irradiation, or usage of thyroxine or any other medications known to affect thyroid hormone levels were excluded. Blood TSH level at the time of diagnosis of thyroid nodule and after were retrospectively studied and analyzed for 100 subjects. Normal blood TSH level range at time of diagnosis was noted to be 0.35-5.50 uIU/mL or 0.35-4.00 uIU/mL depending on the assay method used. Normal repeat blood TSH level range was noted to be 0.35-5.50 uIU/mL, 0.35-4.00 uIU/mL, or 0.55-4.78 uIU/mL depending on the assay method used. Results: 100 subjects with normal blood TSH at the time of diagnosis of their thyroid nodule had blood TSH level monitoring for an average time period of 5.74 years. Out of the 100, six subjects (6%, 95% CI [2.52%, 12.73%]) developed an abnormal blood TSH level in a mean time period of 6.94 years. One subject’s initial blood TSH level increased from 3.38 uIU/mL to 7.76 uIU/mL after 8.52 years. This subject did not receive treatment with thyroxine. In the other five subjects, blood TSH level decreased to below normal range in a mean time period of 6.63 years and ranged from 0.25-0.52 uIU/mL. Blood TSH level remained within normal range for 94 subjects in a mean time period of 5.67 years. Conclusion: Monitoring of blood TSH level over time may not be required in most patients with thyroid nodules in the absence of symptoms, medical conditions, or medications that are known to affect thyroid hormone levels. Presentation Date: Saturday, June 17, 2023

OR32-04 The Usefulness Of The Contrast-enhanced Ultrasound (CEUS) In Evaluating The Risk Of Malignancy In Thyroid Nodules

Abstract Disclosure: A. Zylka: None. K. Dobruch-Sobczak: None. M. Jedrzejczyk: None. H. Piotrzkowska-Wroblewska: None. P. Goralski: None. J. Galczynski: None. E. Bakula-Zalewska: None. M. Dedecjus: None. Objectives Ultrasonography (US) is a primary method used in the evaluation of the thyroid nodules, but there is no single feature predicting malignancy preoperatively with high accuracy. Contrast-enhanced ultrasound (CEUS) provides information about the vascular perfusion and hemodynamics in the thyroid nodules. Therefore, the aim of this paper was to evaluate the usefulness of CEUS in the differential diagnosis of thyroid nodules. Methods Between the years 2021 and 2022, 182 adult patients (149 women and 33 men, aged 18-85) underwent CEUS of a thyroid nodule classified after fine-needle aspiration biopsy with II-VI Bethesda category. During CEUS examination 1.5 ml of SonoVue contrast has been injected and followed by analysing of qualitative CEUS enhancement patterns. All patients subsequently underwent thyroid operation and histopathologic results were used to confirm or exclude the risk of malignancy. Analysis of the visual CEUS enhancement patterns was performed by two experienced specialists in thyroid ultrasonography. Fifteen CEUS qualitative parameters of nodules were estimated and compared with normal thyroid parenchyma including: ring enhancement; low-, iso- and high enhancement; homo- or heterogenous enhancement; slow, fast or equal wash-in and wash-out and centripetal, centrifugal or diffuse enhancement. The statistical analysis of the obtained results was performed using the logistic regression. First sensitivity, specificity and AUC were calculated only for CEUS qualitative features. Then the model was extended using the results based on US B-mode imaging which involved three suspicious features: irregular margins, microcalcifications and marked hypoechogenicity. Results Histopathologic results included 61 benign thyroid nodules (in 51 women and 10 men) and 121 thyroid carcinomas (in 99 women and 22 men). The analysis has pointed five CEUS parameters statistically significant in differential diagnosis between benign and malignant thyroid nodules: low, homogenous and centrifugal enhancement, ring enhancement and slow-wash-out of the contrast. The model, using the above parameters evaluated the risk of malignancy of thyroid nodules with sensitivity of 85.12% and specificity of 45.90% with 0.75 AUC. Including the B-mode features, the sensitivity, specificity and AUC were 85.95%, 62.30% and 0.86, respectively. Conclusions Preliminary results indicate that CEUS is an useful tool in the evaluation of the risk of malignancy in thyroid nodules. The combination of the qualitative enhancement parameters and US B-mode sonographic features increases the usefulness of the method in differential diagnosis between benign and malignant nodules. However, further multicenter studies should be performed in order to standardize qualitative and quantitative CEUS features. Presentation: Sunday, June 18, 2023

THU540 Cribriform Morular Thyroid Carcinoma In A Young Female With Prior Exposure To Whole-Body Radiation For Metastatic Wilms’ Tumor

Abstract Disclosure: D. Caesario: None. M.S. Richardson: None. D. Carneiro-Pla: None. J.K. Fernandes: None. Introduction: Cribriform morular thyroid carcinoma (CMTC) is a rare thyroid cancer that is classically associated with familial adenomatous polyposis (FAP), but can also occur sporadically. We report a case of CMTC in a young female with history of whole-body radiation therapy (WBRT). Case Report: A 19-year-old female with past medical history of Wilms’ tumor (WT) with lung metastasis status post left nephrectomy and WBRT at the age of 6 was seen for newly diagnosed thyroid nodules. Physical exam was remarkable for thyromegaly. Thyroid function tests were normal. Thyroid ultrasound revealed 2 dominant TI-RADS 4 nodules measuring 1.9 x 1.7 x 2.1 cm on the left lobe and 0.9 x 0.8 x 0.8 cm on the right lobe. No suspicious lymphadenopathy. Fine needle aspiration biopsy on both nodules showed findings consistent with papillary thyroid carcinoma (PTC). No family history of thyroid disorders, malignancies, or FAP. She underwent total thyroidectomy with central neck node sampling. Thyroid gland surgical pathology result revealed CMTC positive for thyroid transcription factor-1 (TTF-1), β-catenin, and cytokeratins 5 and 6. It was negative for calcitonin. There was no evidence of invasion, extension, or lymph node metastasis. Genetic tests are currently pending. Discussion: CMTC was previously considered a variant of PTC, accounting for 0.2 to 6% of PTC cases. However, recent observations have challenged this view, and latest WHO update reclassified it as a distinct form of thyroid cancer. It is associated with an indolent clinical course and good prognosis. It almost exclusively affects young women. Both familial and sporadic cases are associated with mutations in the adenomatous polyposis coli (APC) gene. The APC protein is a negative regulator that controls β-catenin concentrations, which is involved in cell adhesion. Mutations in the APC gene may result in FAP and colorectal cancers. In cases associated with FAP, the tumors are usually multifocal and/or bilateral, while sporadic cases usually appear as single nodule. Tumor cells are always positive for TTF-1 and negative for calcitonin. They can also be positive for multiple cytokeratins, but strong nuclear and cytoplasmic staining for β-catenin is the hallmark of this tumor type. The case above occurred in a young female with no significant family history, but who underwent WBRT for WT as a child. Radiation is a known predisposing factor for thyroid cancer, and limited studies on early exposure to radiation revealed higher incidence of sporadic CMTC. However, even among FAP associated cases, CMTC diagnosis preceded that of FAP in up to 40% of the cases. Additionally, up to 20% of WT cases have mutations of the WT1 gene, and among these, at least half also carry acquired somatic mutations in CTNNB1, the gene encoding β-catenin. So far, there has been no reported case of CMTC in the background of WT. It will be interesting to see if she has these underlying mutations. Presentation: Thursday, June 15, 2023

Retracted: The Application Value of SMI Technology and Contrast-Enhanced Ultrasound in the Differential Diagnosis of Benign and Malignant Thyroid Nodules.

[This retracts the article DOI: 10.1155/2022/6157026.].

AI diagnosis of Bethesda category IV thyroid nodules

Thyroid nodules are a common disease, and fine needle aspiration cytology (FNAC) is the primary method to assess their malignancy. For the diagnosis of follicular thyroid nodules, however, FNAC has limitations. FNAC can classify them only as Bethesda IV nodules, leaving their exact malignant status and pathological type undetermined. This imprecise diagnosis creates difficulties in selecting the follow-up treatment. In this retrospective study, we collected ultrasound (US) image data of Bethesda IV thyroid nodules from 2006 to 2022 from five hospitals. Then, US image-based artificial intelligence (AI) models were trained to identify the specific category of Bethesda IV thyroid nodules. We tested the models using two independent datasets, and the best AI model achieved an area under the curve (AUC) between 0.90 and 0.95, demonstrating its potential value for clinical application. Our research findings indicate that AI could change the diagnosis and management process of Bethesda IV thyroid nodules.

Evaluation of elastosonography and fine-needle aspiration biopsy results in the diagnosis of thyroid nodules

Objective The aim of our study is to investigate the importance of strain sonoelastography (SSE) in the diagnosis of thyroid nodules and in the selection of indications for fine-needle aspiration biopsy (FNAB). Materials and methods We included 481 patients who visited our clinic between 2019–2022 with thyroid nodules in the study and used a stepwise diagnostic approach to choose patients to perform surgery on and follow-up. All patients were ranked into 2 groups: Group 1 included 135 (28,1%) patients were assessed as SSE along with traditional examinations. The 2nd group included 346 (71.9%) patients who underwent ultrasound examination using SE, only the hard nodule or hard regions of the nodule detected by SSE were obtained with FNAB. We followed up on nodules with benign FNAB results, and nodules with suspicious or malignant results underwent thyroidectomy. Histopathology results of surgically removed nodules were compared with SSE and FNAB results. AUROC was the primary, while sensitivity, specificity, PPV and NPV were the secondary outcome. Results Out of 481 patients, 420 (87.3%) were female, 61 (12.7%) were male, and the age range was 17–83 (CI=45.57±25.9). The mean nodule size was 23.33±16.7 mm. After evaluating elastography results, FNAB was performed on 346 (71.9%) nodules. Surgical treatment was performed in 114 (group 1) (84.4%) of 135 patients, benign tumors were diagnosed in 107 (93.8%), and malignant tumors in 7 (6.2%). Surgical treatment in group 2 was performed in 146 (42.2%) of 346 patients, benign tumors were detected in 105 (71.9%) patients, and malignant in 41 (28.9%). The combined use of the sonoelastography and fine-needle aspiration biopsy contributed to a significant increase in the frequency of detection of thyroid nodules of malignant structure, which amounted to 28.9% in patients of the main group (versus 6.2% in patients of the control group) and a decrease in the frequency of surgical interventions by more than 2 times in patients of the second group. The area of the ROC curve for SSE was S=0.851±0.029; 95% CI (0.794 to 0.908); P=0.000. It is a statistically significant marker in the diagnosis of thyroid nodules. The most important marker in the diagnosis of thyroid nodules was the FNAB test: S=0.874±0.034; 95% CI (0.807 to 0.940); P =0.000. Sensitivity, specificity, PPV, and NPV of elastosonography of the I group were 42.9±18.7%, 96.3±1.8%, 42.9%, 96.3; of II group 68.3±7.3%, 87.6±3.2%, 68.3%, 87.6%; FNAB 90.2±4.6%, 90.5±2.9%, 76.2%, 91.3%, respectively. Conclusion Combined use of SE and Bethesda system was able to detect malignancy in the early stages and to reduce the number of operations.

TS-DSANN: Texture and shape focused dual-stream attention neural network for benign-malignant diagnosis of thyroid nodules in ultrasound images.

Recently, accurate diagnosis of thyroid nodules has played a critical role in precision medicine and healthcare system management. Due to complicated changes in ultrasound features of texture, and similar visual appearance of benign-malignant nodules, the identification of cancerous thyroid lesions from a given ultrasound image still faces challenges for even experienced radiologists. Learning-based computer-aided diagnosis (CAD) systems have accordingly attracted more and more attention recently. However, little research is committed to developing a deep learning-based CAD system that has greater conformity with radiologists' diagnostic decision-making. In this study, we devise a texture and shape focused dual-stream attention neural network, dubbed TS-DSANN. Specifically, in the texture focused stream, we utilize the ImageNet pre-trained ResNet34 to guide the network to recognize texture-related nodule attributes. Meanwhile, in the shape focused stream, in addition to using ResNet34 backbone, jointly learning from scratch with the contour obtained by contour detection module to enhance the extraction of shape features. Afterward, we employ a concatenation operation to aggregate the abovementioned two stream networks for capturing richer and more representative features. Finally, we further utilize an online class activation mapping mechanism to assist the dual-stream network in generating a localization heatmap to obtain more visualization attention to the nodule from the whole image, and supervise classifier's attention in decision-making. Experimental results conducted on the two-center thyroid nodule ultrasound datasets verify that our proposed method has improved the classification performance, superior to the state-of-the-art methods.