A lightweight network for automatic thyroid nodules location and recognition with high speed and accuracy in ultrasound images

BACKGROUND

The intelligent diagnosis of thyroid nodules in ultrasound image is an important research issue. Automatically locating the region of interest (ROI) of thyroid nodules and providing pre-diagnosis results can help doctors to diagnose faster and more accurate.

OBJECTIVES

This study aims to propose a model, which can detect multiple nodules stably and accurately in order to avoid missed detection and misjudgment. In addition, the detection speed of the model needs to be fast for real-time diagnosis in ultrasound images.

METHODS

Based on the object detection technology, we propose an accurate, robust and high-speed network with multiscale fusion strategy called Efficient-YOLO, which can realize the localization and recognition of nodules at the same time. Finally, multiple metrics are used to measure the diagnostic ability of the model.

RESULTS

Experimental results conducted on 3,562 ultrasound images show that our new model greatly increases the accuracy and speed of the detection compared with the baseline model. The best mAP is 92.64%, and the fastest detection speed is 45.1 frames per second.

CONCLUSIONS

This study proposed an effective method to diagnosis thyroid nodules automatically, which can meet the real-time requirements, indicating that its effectiveness and feasibility for future clinical application.

Benign and malignant thyroid nodules with autoimmune thyroiditis

Objective

The prevalence of autoimmune thyroiditis (AT) in papillary thyroid carcinoma (PTC) is still controversial. The aim of this study was to investigate the frequency of coexistence of PTC with AT versus that of the coexistence of benign nodules with AT.

Materials and methods

This was a cross-sectional retrospective study including patients operated on for thyroid nodules from January 2011, to April 2021. The frequency of papillary carcinomas cooccurring with AT was compared to that of benign nodules cooccurring with AT, which was assessed based on cytopathological diagnosis after thyroidectomy.

Results

The study included 668 cases of benign nodules and 420 cases with PTC. No statistically significant difference was observed between cases of benign and PTC nodules regarding the presence of AT (25% vs. 28%, respectively, p = 0.177). The size of the PTC compared to that of the benign predominant nodules was significantly smaller both in the absence (0.96 ± 1.09 cm vs. 2.19 ± 1.06 cm, p < 0.05) and in the presence (0.77 ± 0.76 cm vs. 1.67 ± 1.08 cm, p < 0.01) of AT. In the binary logistic regression analysis of the PTC, the only variable associated with AT was multifocality (odds ratio: 1.750, 95% confidence intervals: 1.131-2.706, p = 0.013). The incidences of lymph node involvement and advanced stage PTC were very low both in the presence and absence of AT.

Conclusion

The nodules present with PTC were not more likely to coexist with AT than benign nodules were. The small incidence of advanced PTC indicates a significant improvement in early-stage diagnosis.

Down-regulation of miR-7-5p and miR-548ar-5p predicts malignancy in indeterminate thyroid nodules negative for BRAF and RAS mutations

PURPOSE

The value of molecular markers in refining preoperative risk assessment of indeterminate thyroid nodules is being widely investigated. MicroRNAs (miRNA) are emerging as promising biomarkers for diagnostic and prognostic purposes. The aim of this study is to identify miRNAs specifically deregulated in mutation-negative indeterminate thyroid nodules.

METHODS

Ninety-eight nodules preoperatively diagnosed as TIR 3A or TIR 3B with available histological diagnosis of follicular adenoma (FA), noninvasive follicular neoplasm with papillary-like nuclear features (NIFTP), and follicular variant papillary thyroid carcinoma (FV-PTC) have been retrospectively selected. Mutations in BRAF and RAS genes have been tested in all samples by real-time PCR; miRNAs were purified from cytology slides of 60 samples; expression analysis of 798 miRNAs was measured by the nCounter system.

RESULTS

Point mutations in BRAF and RAS genes were detected in 32 out of 98 nodules (32.7%), the majority of which in FV-PTCs. Differential expression of miRNA in wild-type nodules highlighted that two miRNAs, namely miR-7-5p and miR-548ar-5p, were downregulated in FV-PTCs compared to FAs. The combined expression of these miRNAs, tested by ROC analysis, showed an area under the curve of 0.79. Sensitivity and negative predictive value were high both in wild-type (93% and 92%, respectively) and in mutated nodules (94% and 85%, respectively).

CONCLUSION

The analysis of miR-7-5p and miR-548ar-5p expression in indeterminate thyroid nodules demonstrated a promising value in ruling out malignancy.

The Future of Thyroid Nodule Risk Stratification

Clinical evidence supports the association of ultrasound features with benign or malignant thyroid nodules and serves as the basis for sonographic stratification of thyroid nodules, according to an estimated thyroid cancer risk. Contemporary guidelines recommend management strategies according to thyroid cancer risk, thyroid nodule size, and the clinical scenario. Yet, reproducible and accurate thyroid nodule risk stratification requires expertise, time, and understanding of the weight different ultrasound features have on thyroid cancer risk. The application of artificial intelligence to overcome these limitations is promising and has the potential to improve the care of patients with thyroid nodules.

Contrast-enhanced ultrasound improves the potency of fine-needle aspiration in thyroid nodules with high inadequate risk

Background

This study aims to determine the clinical value of contrast enhanced ultrasound (CEUS) for fine-needle aspiration (FNA) of high inadequate risky thyroid nodules.

Methods

During April 2018 and April 2021, consecutive 3748 thyroid nodules underwent FNA were retrospectively analyzed. CEUS guided FNA (CEUS-FNA) was applied in 115 nodules with high inadequate risk in Lingnan Campus. Ten nodules underwent CEUS-FNA presented non-enhancing, and would be further analyzed independently. Other 105 partial or total enhancing nodules were included as CEUS-FNA group, and 210 nodules with high inadequate risk in Tianhe Campus were match as the US-FNA control group. FNA specimens were collected for liquid-based preparation. Cytological results were classified following the Bethesda System for Reporting Thyroid Cytopathology.

Results

The overall FNA specimen inadequate rate in our center was 6.6%. All of the ten non-enhancing nodules under CEUS have an inadequate result in cytopathological analyzes. The subsequent postoperative pathology and follow-up ultrasonography showed the non-enhancing nodules were benign or stable without further malignant features. Total specimen inadequate rate of high inadequate risk thyroid nodules in CEUS-FNA group was significantly lower than US-FNA group (6.7% vs. 16.7%, P = 0.014). Further stratified analyzed shown that FNA under US guidance, the inadequate rates in cystic, predominantly cystic, predominantly solid and solid sub-groups were 28.1%, 17.1%, 10.0% and 9.2% (P = 0.019). In contrast, the inadequate rates in cystic, predominantly cystic, predominantly solid and solid sub-groups were 7.4%, 6.7%, 5.6% and 6.7% (P = 0.996) in CEUS-FNA group.

Conclusions

CEUS can improve the specimen adequacy of FNA in high inadequate risk thyroid nodules by avoiding unnecessary FNAs of the non-enhancing nodules, and accurately locating the viable tissue and precise guidance in real-time. CEUS is a recommend modality for FNA guidance of high inadequate risk thyroid nodules.

Application value of gray-scale ultrasound and shear wave elastography in differential diagnosis of thyroid nodules

BACKGROUND

The global prevalence of thyroid cancer has increased significantly in recent years. Ultrasonography is the preferred method for differentiating benign and malignant thyroid nodules preoperatively and is recommended by guidelines.

OBJECTIVE

To assess the application value of gray-scale ultrasound and shear wave elastography in distinguishing small thyroid nodules.

METHODS

A retrospective analysis of 228 thyroid nodules, all of which were confirmed by pathology after surgery or FNA from January 2019 to January 2020, was carried out. All nodules were divided into a ⩽ 5 mm group and a > 5 mm group according to their maximum size. We compared the differences in the gray scale and elastography of the nodules between the two groups and the accuracy of different diagnostic methods.

RESULTS

The accuracies of gray-scale ultrasound and shear wave elastography in the ⩽ 5 mm group were found to be lower than those in the > 5 mm group, and the gray-scale accuracy was slightly higher than that of shear wave elastography in both groups (p< 0.05). The largest AUC (area under the curve) of elastic parameters in the ⩽ 5 mm and > 5 mm groups was found for Emax and Esd, respectively. Based on a combination of these two parameters, the accuracies of the two groups were significantly higher than those of the parameters or gray scale alone (p< 0.05) and were 84.62% and 85.48%, respectively.

CONCLUSION

Shear wave elastography is valuable in the diagnosis of benign and malignant thyroid nodules using ultrasonography. When combining gray-scale ultrasound and shear wave elastography, the diagnostic accuracy is obviously improved, especially for ⩽ 5 mm small thyroid nodules.

Quality Improvement in Thyroid Fine-Needle Aspiration Biopsy accuracy at San Juan City Hospital Endocrinology Clinics

OBJECTIVE

This study aimed to determine if there was a reduction in the amount of non-diagnostic cytopathology results of thyroid fine-needle aspiration (FNA) biopsies performed at San Juan City Hospital (SJCH) endocrinology clinics since the inclusion of a cytopathologist at the clinics.

METHODS

This research consisted of a retrospective analysis of thyroid nodule FNA biopsy results performed at SJCH endocrinology clinics. The biopsies analyzed were performed during academic years from July 2017-June 2018 and July 2018-June 2019, a period that reflects one academic year prior and a year after the inclusion of a cytopathologist to the clinics. The patients were classified into "pre group" and "post -group." Descriptive analysis was conducted, taking into consideration variables including sex, age, period, location, size of the nodule, and cytology results. A Chi-square test and Confidence Interval were used to assess the association and estimates between predictors and outcomes.

RESULTS

From the 145 thyroid nodules biopsied, a total of 121 nodules (83.4%) resulted in diagnostic cytologic results, while 24 nodules (16.6%) were non-diagnostic. From the "pre group," 57 nodules (78.1%) had a diagnosis, while the other 16 (21.9%) were reported as non-diagnostic. From the "post group", 64 nodules (88.9%) had a diagnosis, while the other 8 (11.1%) resulted in non-diagnostic findings (p-value: = 0.08). Even though results were statistically non-significant, a clear trend towards a decrease in non-diagnostic samples was evident.

CONCLUSION

In our study, there was a decrease in the number of non-diagnostic thyroid nodule FNA results after on-site adequacy determination guided by a cytopathologist.

Incidence of thyroid nodules in early stage autosomal polycystic kidney disease

Background

Autosomal dominant polycystic kidney disease (ADPKD) is the most common hereditary kidney disease. Defect in cilia-mediated signaling activity is a crucial factor leading to cyst formation. Hence, ADPKD is regarded as a systemic disorder with multiple extrarenal complications, including cysts in other organs, for instance, the liver, pancreas, spleen, or ovaries. Interestingly, loss-of-function of primary cilia has been recently found to contribute to a malignant transformation from degenerated thyroid follicles. However, the increased incidence of thyroid nodules in ADPKD patients has not yet been fully confirmed.

Objectives

To determine the incidence of thyroid lesions in patients with ADPKD in comparison to previous population studies. Moreover, we aimed to investigate if the pace of the disease progression is associated with a higher prevalence of thyroid lesions.

Material and methods

In 49 early-stage ADPKD patients recruited from our center, we performed ultrasonography of the thyroid glands, and laboratory evaluation of thyroids function. We compared the results with population studies.

Results

Twenty-three individuals had solid, cystic-solid, or cystic lesions revealed in the ultrasonography and 2 patients had a positive past medical history for thyroidectomy due to nodular goiter. In 10 patients out of the 23, only minor cysts with no clinical significance were found and 13 out of the 23 patients had solid or cystic-solid lesions, which occurred to be benign based on three years of follow-up or the biopsy of the nodule.

Conclusions

We found no increased incidence of thyroid gland lesions in early ADPKD patients in comparison to previous population studies. Plausibly, mechanisms other than defective cilia signaling are involved in the risk for focal thyroid lesions formation. Moreover, the rate of progression of kidney function decline seems to be not accompanied by the higher incidence of thyroid pathology.

Adaptive soft erasure with edge self-attention for weakly supervised semantic segmentation: Thyroid ultrasound image case study

[S U M M A R Y] Weakly supervised segmentation for medical images ease the reliance of models on pixel-level annotation while advancing the field of computer-aided diagnosis. However, the differences in nodule size in thyroid ultrasound images and the limitations of class activation maps in weakly supervised segmentation methods typically lead to under- and/or over-segmentation problems in real predictions. To alleviate this problem, we propose a weakly supervised segmentation neural network approach. This new method is based on a dual branch soft erase module that expands the foreground response region while constraining the erroneous expansion of the foreground region by the enhancement of background features. The sensitivity of this neural network to the nodule scale size is further enhanced by the scale feature adaptation module, which in turn generates integral and high-quality segmentation masks. In addition, while the nodule area can be significantly expanded through soft erase module and scale feature adaptation module, the activation effect in the nodule edge area is still not satisfactory, so that we further add an edge-based attention mechanism to strengthen the nodule edge segmentation effect. The results of experiments performed on the thyroid ultrasound image dataset showed that our new approach significantly outperformed existing weakly supervised semantic segmentation methods, e.g., 5.9% and 6.3% more accurate than the second-based results in terms of Jaccard and Dice coefficients, respectively.

TNSNet: Thyroid nodule segmentation in ultrasound imaging using soft shape supervision

BACKGROUND AND OBJECTIVES

Thyroid nodules are a common disorder of the endocrine system. Segmentation of thyroid nodules on ultrasound images is an important step in the evaluation and diagnosis of nodules and an initial step in computer-aided diagnostic systems. The accuracy and consistency of segmentation remain a challenge due to the low contrast, speckle noise, and low resolution of ultrasound images. Therefore, the study of deep learning-based algorithms for thyroid nodule segmentation is important. This study utilizes soft shape supervision to improve the performance of detection and segmentation of boundaries of nodules. Soft shape supervision can emphasize the boundary features and assist the network in segmenting nodules accurately.

METHODS

We propose a dual-path convolution neural network, including region and shape paths, which use DeepLabV3+ as the backbone. Soft shape supervision blocks are inserted between the two paths to implement cross-path attention mechanisms. The blocks enhance the representation of shape features and add them to the region path as auxiliary information. Thus, the network can accurately detect and segment thyroid nodules.

RESULTS

We collect 3786 ultrasound images of thyroid nodules to train and test our network. Compared with the ground truth, the test results achieve an accuracy of 95.81% and a DSC of 85.33. The visualization results also suggest that the network has learned clear and accurate boundaries of the nodules. The evaluation metrics and visualization results demonstrate the superior segmentation performance of the network to other classical deep learning-based networks.

CONCLUSIONS

The proposed dual-path network can accurately realize automatic segmentation of thyroid nodules on ultrasound images. It can also be used as an initial step in computer-aided diagnosis. It shows superior performance to other classical methods and demonstrates the potential for accurate segmentation of nodules in clinical applications.

Efficacy in size and symptom reduction of radiofrequency ablation of benign non-functioning thyroid nodules

INTRODUCTION

Radiofrequency ablation (RFA) plays an increasing role in the management of thyroid nodules. The purpose of this study was to evaluate the safety and efficacy of RFA of benign thyroid nodules in terms of volume reduction and symptom relief.

METHODS

We reviewed the medical records of patients with a solitary thyroid nodule who received RFA at our centre from April 2016 to January 2020. Ultrasound-guided RFA was performed by the moving-shot technique. Patients were followed up with ultrasound examinations and thyroid function tests. We evaluated changes in volume and in compressive (0-10 scale) and cosmetic (4-point scale) symptoms to analyse efficacy and complications, and in thyroid function to evaluate safety.

RESULTS

A total of 72 patients were included in the study. The mean follow-up was 11.1±0.7 months. There was a statistically significant reduction in size of the nodules at Month 1, 3, 6 and 12 after RFA (expressed as volume reduction rate: 34%±2.1, 50.8%±2.2, 60.3%±2.6, 58.87%±3.5, p<0.0001). We found statistically significant improvement in compressive symptoms (from 7.1±0.26 to 1.76±0.33, p<0.0001) and in cosmetic alteration (from 3.66±0.09 to 2.14±0.14, p<0.0001). No major complications were observed.

DISCUSSION

RFA achieved significant volume reduction and improved compressive symptoms as well as cosmetic complaints, with minimum side effects. Our results are in line with recent available evidence and further support the use of RFA as a safe and effective therapeutic option in the management of benign thyroid nodules.

Stratification of follicular thyroid tumors using data-independent acquisition proteomics and a comprehensive thyroid tissue spectral library

Thyroid nodules occur in about 60% of the population. A major challenge in thyroid nodule diagnosis is to distinguish between follicular adenoma (FA) and carcinoma (FTC). Here, we present a comprehensive thyroid spectral library covering five types of thyroid tissues. This library includes 121 960 peptides and 9941 protein groups. This spectral library can be used to quantify up to 7863 proteins from thyroid tissues, and can also be used to develop parallel reaction monitoring (PRM) assays for targeted protein quantification. Next, to stratify follicular thyroid tumours, we compared the proteomes of 24 FA and 22 FTC samples, and identified 204 differentially expressed proteins (DEPs). Our data suggest altered ferroptosis pathways in malignant follicular carcinoma. In all, 31 selected proteins effectively distinguished follicular tumours. Of those DEPs, nine proteins were further verified by PRM in an independent cohort of 18 FA and 19 FTC. Together, we present a comprehensive spectral library for DIA and targeted proteomics analysis of thyroid tissue specimens, and identified nine proteins that could potentially distinguish FA and FTC.

[Investigate the thyroid function of radiation workers and analysis of influence factors]

Objective: To analyze the thyroid condition and influencing factors of radiation workers, and to provide scientific basis for radiation protection management. Methods: In April 2020, 4308 radiation workers from October 1, 2017 to September 30, 2019 were collected for occupational health examination in Hangzhou Occupational Disease Prevention and Treatment Hospital, the data included basic information, Thyroid hormone level and thyroid color doppler ultrasound results. The thyroid status of radiation workers in different subgroups was compared. The influencing factors of Thyroid nodule were analyzed by multiple logistic regression. Results: 616 radiation workers had abnormal thyroid gland (14.3%, 616/4308) . The main manifestations of thyroid gland abnormality were abnormal Thyroid nodule (5.1%, 220/4308) and abnormal TSH level (7.1%, 308/4308) . Compared with the male radiation workers, the abnormal rate of Thyroid nodule, T3 and Tsh in female radiation workers was higher (P<0.05) , and the abnormal rate of Thyroid nodule, T3 and T4 increased with the increase of working age (P<0.05) . Radiation Workers in non-rated medical institutions and private medical institutions had the highest detection rate of Thyroid nodule (P<0.05) , and there was no significant difference in thyroid abnormalities among Workers (P<0.05) . By multiple logistic regression analysis, Sex (female) , age and institution type (private) were all independent risk factors for Thyroid nodule (95% CI: 1.548~2.763、1.002~1.030, P<0.05) . Conclusion: Ionizing radiation can cause thyroid damage in radiation workers, so we should pay more attention to the radiation protection management of female, high age, private and district medical organizations, so as to protect the health of radiation workers.

Testing of mutations on thyroid nodules with indeterminate cytology: A prospective study of 112 patients in Argentina

BACKGROUND

The study of genetic mutations in thyroid nodules makes it possible to improve the preoperative diagnosis of and reduce unnecessary surgeries on benign nodules. In this study, we analysed the impact of implementing a 7-gene mutation panel that enables mutations to be detected in BRAF and RAS (H/N/K) and the gene fusions PAX8/PPARG, RET/PTC1 and RET/PTC2, in a population in northern Argentina.

METHODS

We performed a prospective analysis of 112 fine needle aspirations diagnosed as having indeterminate cytology according to the Bethesda classification system. These include the Bethesda III or atypia of unknown significance/follicular lesion of unknown significance and Bethesda IV or follicular neoplasm/suspicious for follicular neoplasm categories. The mutations of the 7-gene panel were analysed and this information was linked to the available histology and ultrasound monitoring.

RESULTS

The BRAF V600E and RET/PTC1 mutations were associated with carcinoma in 100% of cases (n = 8), whereas only 37.5% (n = 3) of the nodules with RAS and 17% (n = 1) with PAX8/PPARG mutations were associated with carcinoma. From the histological diagnosis and ultrasound monitoring of patients, we can estimate that this panel has a sensitivity of 86% in detecting malignant carcinoma, a specificity of 77%, a positive predictive value (PPV) of 54% and a negative predictive value (NPV) of 94%. In this study, it was possible to reduce the number of surgeries by 48% in the patients analysed.

CONCLUSION

The implementation of the mutation panel allowed the appropriate surgical strategy to be selected for each patient, the number of two-step surgeries to be reduced, and active follow-up to be established in low-risk patients.

Use of Web-Based Calculator for the Implementation of ACR TI-RADS Risk-Stratification System

In this article, we demonstrate the use of a software-based radiologist reporting tool for the implementation of American College of Radiology Thyroid Imaging, Reporting and Data System thyroid nodule risk-stratification. The technical details are described with emphasis on addressing the information security and patient privacy issues while allowing it to integrate with the electronic health record and radiology reporting dictation software. Its practical implementation is assessed in a quality improvement project in which guideline adherence and recommendation congruence were measured pre and post implementation. The descriptions of our solution and the release of the open-sourced codes may be helpful in future implementation of similar web-based calculators.

The relationship between ultrasound-based TIRADS and BETHESDA categories in patients undergoing thyroid biopsy

The TIRADS is a scoring system used for the selection of nodules for FNA and classification of the risk of malignancy based on ultrasound characteristics. The BETHESDA is a standard reporting system used for the classification of FNA results based on six criteria with risks for malignancy. The objective of this study was to evaluate the relationship between TIRADS and BSRTC classifications in patients undergoing thyroid biopsy. A total of 350 consecutive patients were retrospectively evaluated using TIRADS and BETHESDA reporting systems for determining preoperative diagnosis of thyroid nodules. Patients' demographics, size, echogenicity and contour status of the nodules, TIRADS and BETHESDA scores were recorded and analyzed. Data obtained in this study were expressed as mean, standard deviation, frequency and percentage descriptive statistics. The mean age of the patients was 49.03 ± 17.58 years. The mean nodule size was measured as 20.56 ± 10.47 mm. TIRADS TR3 category was found in 165 (47.14%), TR4 in 154 (44%) and TR5 in 31 (8.86%) patients, while BETHESDA II category was found in 288 (82.28%), BETHESDA III category in 1 (0.29%), BETHESDA IV category in 19 (5.43%), BETHESDA V in 37 (10.57%) and BETHESDA VI in 5 (1.43%) patients. There was a general concordance between BETHESDA and TRIADS categories. The most significant concordance was found between BETHESDA IV and TR4 categories (84.21%). Combined use of TRIADS and BETHESDA can be efficiently used to provide the most accurate results for making preoperative diagnosis of thyroid nodules and to determine the risk of malignancy.

Fine-Needle Aspiration Under Guidance of Ultrasound Examination of Thyroid Lesions

Fine-needle aspiration biopsy is the most common method for preoperative diagnosis of thyroid carcinomas including papillary carcinoma. The procedure is best performed with ultrasound by operator with professional skill and knowledge. Several guidelines recommend the indication of fine-needle aspiration concerning the pattern of ultrasound and size of nodules. Besides, fine-needle aspiration biopsy of lymph nodes should be performed if malignancies are suspected. Fine-needle aspiration biopsy of thyroid gland is mostly safe, but complications such as blood extravasation-related complications, acute thyroid enlargement, infection in thyroid gland, and pneumothorax could occur. The most frequent complications are blood extravasation-related complications, which could be fatal. Similarly, acute thyroid enlargement could also be severe. To conclude, fine-needle aspiration biopsy is useful and should be performed under the precise indication and the updated knowledge of complications including the way of handling if they occur.

Observation Variation in Ultrasonography Assessment of Thyroid Nodules

OBJECTIVES

Accurate detection and competent management of thyroid nodules, as a common disease, basically depends on the reliability of the ultrasonography (US) report. In this research, we evaluated inter and intra-observer variation among ultrasonography reporters, based on ACR-TIRADS.

METHODS

In this retrospective study, 345 thyroid US images of 150 patients were reviewed. Three clinicians with at least 6-year experience in thyroid US reviewed the images twice at 6-8 weeks' intervals. Composition, echogenicity, shape, margin, and echogenic foci based on ACR-TIRADS were reported, independently. Inter and intra-observer variations were calculated based on Cohen's Kappa statistics.

RESULTS

345 ultrasonography images of 150 patients with thyroid nodules (83 women and 67 men) with a mean age of 65 years were reviewed. Moderate to the substantial intra-observer agreement was achieved with the highest Kapa value in the category of shape (k=0.61-0.77). For TIRADS level, the moderate intra-observer agreement was observed (k=0.42-0.46). Inter-observer agreement for the US category of thyroid nodules was obtained slightly to moderate. Composition (k=0.42 and 0.51) and echogenicity (k=0.45 and 0.46) showed the highest overall agreement and margin showed the lowest overall agreement (k=0.18 and 0.19). In assessing TIRADS level of nodules, a fair agreement was obtained (k=0.23 and 0.29).

CONCLUSION

Moderate to substantial intra-observer agreement and slight to moderate inter-observer variation for evaluation of thyroid nodules; shows the need for a computer-aided diagnosis system based on artificial intelligence to assist our physicians in differentiating thyroid nodule characteristics based on explicit image features. An additional training course based on ACR-TIRADS for physicians can be another useful recommendation.

Neutrophils to lymphocytes ratio between benign and malignant thyroid nodule

Objectives:

To determine the relationship of Neutrophils to lymphocytes ratio (NLR) values with the benign and malignant thyroid nodules.

Methods:

In this cross-sectional clinical study conducted from September 1^st 2020 to February 28^th 2021, we included 216 patients who underwent thyroidectomy at Jinnah Postgraduate Medical Centre Karachi Pakistan. After thyroidectomy specimens were sent for pathologic examination. Patients were divided into two categories based on histopathologic results; Malignant nodule and benign nodules. Data of complete blood count was obtained from the pre-operative lab investigations and NLR was calculated.

Results:

There were 42 (26%) men, and 116 (74%) women of 158 in the BTN group, 18 (31%) men, and 40 (69%) women of 58 in the MTN group. The mean age of 48 ± 6 years in the BTN group as well as 47 ±8 years in the MTN group (p-value 0.32). The mean neutrophil count in the BTN group was 4.26 ± 2.8 versus 4.41 ± 2.2 (x 1000/mm³)) the malignance thyroid group (p-value = 0.71). The mean lymphocyte count was 3.81 ± 0.9(x 1000/mm³) in the BTN group and 3.61± 1.2 (x 1000/mm³) in the malignance group (p-value = 0.18). The mean NLR value for the benign thyroid nodular group was 1.19 ± 2.2 and 1.22±1.8 in the malignant thyroid nodular group (p-value = 0.92).

Conclusion:

According to the results of this study, we concluded that preoperative period biochemistry laboratory results such as neutrophils count, lymphocyte count, and NLR value don’t provide enough evidence to differentiate between benign and malignant thyroid carcinoma.

Current Iodine Nutrition Status and Morbidity of Thyroid Nodules in Mainland China in the Past 20 Years

The aim of this study was to comprehensively assess the prevalence of goiter and thyroid nodules (TNs) in relation to China's iodine nutrition level over the past 20 years and provide an effective reference for developing health policies. PubMed, EMBASE, Chinese National Knowledge Infrastructure, Chongqing VIP, and Chinese Wan Fang databases were searched for relevant studies from Jan 1996 to Feb 2020. Two reviewers extracted valid data from the eligible citations to determine the morbidity of TNs in different urinary iodine concentrations (UICs) and in patients of different genders, of different ages, who live in different geographic regions, and who live at different altitudes, as well as the P values of interactions between groups. There were 26 articles (34 studies) included in this analysis. The overall morbidity of TNs in mainland China was 23.4%. Morbidity was higher in urban areas (P < 0.001) than in rural and mixed areas. Coastal areas (P < 0.001), female patients (P < 0.001), high-altitude areas (P < 0.001), and residence in south China (P < 0.001) were all associated with higher morbidity of TNs. The lowest morbidity value of TNs, 16%, was in the more-than-adequate iodine subgroup. The highest morbidity, 27.2%, was in the adequate iodine subgroup. The morbidity of TNs increases with age, and women are more likely to have TNs. We also need to perform more epidemiological studies, and in the future, we should cultivate better understanding of the relationship between other thyroid diseases and provide more comprehensive and useful information for other researchers.

Strain Imaging in the Evaluation of Thyroid Nodules: The Associated Factors Leading to Misdiagnosis

This study analyzed which imaging characteristics may contribute to misdiagnosis of benign and malignant thyroid nodules (TNs) through strain imaging (SI). We analyzed SI parameters and conventional ultrasound (CUS) characteristics of 264 nodules with histopathology results and we applied binary logistic regression analysis to select independent variables for incorrect SI results. The elastography imaging (EI) score and the virtual touch tissue imaging (VTI) score of malignant TNs were remarkably higher than that in benign TNs. VTI, of which the false-positive rate (FPR), the false-negative rate (FNR), and the cutoff value were 47.6% (39/82), 17.6% (32/182) and 3.5, presented higher diagnostic performance than EI. Aspect ratio (odd ratio [OR]: 14.208) and intra-nodular calcification (OR: 20.384) were significantly associated with false-positive VTI findings, while posterior acoustic feature (OR: 30.424) was significantly associated with false-negative VTI findings. Sex (OR: 0.221) and aspect ratio (OR: 0.268) were negatively associated with false-negative EI findings. In conclusion, sex and aspect ratio significantly affect the accuracy of EI results, while aspect ratio, intra-nodular calcification and posterior acoustic features notably affect the accuracy of VTI results. These CUS features of TNs should not be ignored before explaining SI examination results.

Large thyroid nodules: should size alone matter?

BACKGROUND

The management of thyroid nodules ≥ 4 cm is controversial due to conflicting evidence on the prevalence of malignancy and diagnostic accuracy of fine-needle aspiration cytology (FNAC). Some literature recommends routine excision of large thyroid nodules due to high cytology false negative rates (FNR). We aim to investigate the diagnostic accuracy of FNAC, prevalence of malignancy in large (≥ 4 cm) thyroid nodules compared to nodules < 4 cm, and the clinical and ultrasound characteristics of those large nodules with false negative cytology.

METHODS

This was a retrospective case-log review in a tertiary referral hospital. All thyroid nodules subjected to Ultrasound (US)-guided FNAC by the Interventional Radiology department between December 2011 and November 2017 were included. Data on patient demographics, thyroid US features, cytology findings, and surgical histology were collected and analyzed. Sensitivity, specificity, and FNR were calculated based on FNAC results and final post-operative histology. Factors associated with a false negative result were analyzed using univariate and multivariate analyses.

RESULTS

A total of 4982 nodules were studied, including 4419 < 4 cm and 563 ≥ 4 cm. Malignancy rates were similar in both groups. For nodules ≥ 4 cm, FNAC sensitivity was 40%, specificity 100%, and FNR 6.6% compared to 4.2% in nodules < 4 cm. Within malignant nodules, there was a significantly higher proportion of follicular and Hurthle cell carcinomas in nodules ≥ 4 cm. Amongst nodules ≥ 4 cm, multivariate analysis revealed male gender to be an independent predictor of FNR (OR 3.32; 95% CI 1.29-8.59).

CONCLUSION

Larger nodules ≥ 4 cm have a similar malignancy rate as nodules < 4 cm, and FNAC FNR is low at 6.6%. Management of large thyroid nodules should be individualized based on their clinical, sonographic and cytological features rather than routine surgery.

LEVEL OF EVIDENCE

3.

Are the anatomical, clinical, and ultrasound characteristics of thyroid nodules with Bethesda III or IV cytology and ACR TI-RADS 3, 4, or 5 able to refine the indications for molecular diagnostic tests?

OBJECTIVE

To analyze the association of clinical, anatomical, and ultrasound (US) characteristics of malignancies in Bethesda III or IV (III-B or IV-B) thyroid nodules.

METHODS

The association between malignancies and the following variables were analyzed: III-B or IV-B, age < 55 years and ≥ 55 years, sex, family history of thyroid cancer, history of irradiation, nodule size, and ACR TI-RADS classification in 62 participants who underwent thyroidectomy.

RESULTS

Of the 62 participants, 87.1% (54/62) were women, 74.2% were < 55 years old, 95.2% had no family history of thyroid cancer, 56.5% had nodules < 2 cm in size, 62.9% were IV-B, and 69.4% were ACR TI-RADS 4. Thirty-two patients had thyroid carcinoma, and 30 had benign histology. Among all factors associated with malignancy, only ACR TI-RADS 5 classification on US was found to be statistically significant (p = 0.014), while III-B with architectural atypia cytological classification was the only one significantly associated with benign status (p = 0.004).

CONCLUSION

Only a high risk of malignancy as assessed using US was able to refine the indication for molecular tests in a group of patients with indeterminate nodules. We found 85% (53/62) of III-B or IV-B thyroid nodules would benefit from available molecular diagnostic tests.

An integrated AI model to improve diagnostic accuracy of ultrasound and output known risk features in suspicious thyroid nodules

OBJECTIVES

From the viewpoint of ultrasound (US) physicians, an ideal thyroid US computer-assisted diagnostic (CAD) system for thyroid cancer should perform well in suspicious thyroid nodules with atypical risk features and be able to output explainable results. This study aims to develop an explainable US CAD model for suspicious thyroid nodules.

METHODS

A total of 2992 solid or almost-solid thyroid nodules were analyzed retrospectively. All nodules had pathological results (1070 malignancies and 1992 benignities) confirmed by ultrasound-guided fine-needle aspiration cytology and histopathology after thyroidectomy. A deep learning model (ResNet50) and a multiple risk features learning ensemble model (XGBoost) were used to train the US images of 2794 thyroid nodules. Then, an integrated AI model was generated by combining both models. The diagnostic accuracies of the three AI models (ResNet50, XGBoost, and the integrated model) were predicted in a testing set including 198 thyroid nodules and compared to the diagnostic efficacy of five ultrasonographers.

RESULTS

The accuracy of the integrated model was 76.77%, while the mean accuracy of the ultrasonographers was 68.38%. Of the risk features, microcalcifications showed the highest contribution to the diagnosis of malignant nodules.

CONCLUSIONS

The integrated AI model in our study can improve the diagnostic accuracy of suspicious thyroid nodules and output the known risk features simultaneously, thus aiding in training young ultrasonographers by linking the explainable results to their clinical experience and advancing the acceptance of AI diagnosis for thyroid cancer in clinical practice.

KEY POINTS

• We developed an artificial intelligence (AI) diagnosis model based on both deep learning and multiple risk feature ensemble learning methods. • The AI diagnosis model showed higher diagnostic accuracy for suspicious thyroid nodules than ultrasonographers. • The AI diagnosis model showed partial explainability by outputting the known risk features, thus aiding young ultrasonic doctors in increasing the diagnostic level for thyroid cancer.