The Diagnostic Efficiency of Ultrasound Computer-Aided Diagnosis in Differentiating Thyroid Nodules: A Systematic Review and Narrative Synthesis

Association of urinary chlorpyrifos, paraquat, and cyproconazole levels with the severity of fatty liver based on MRI

BACKGROUND

The objective of this study was to detect the urinary levels of chlorpyrifos, paraquat, and cyproconazole in residents living in Fuyang City and to analyze the correlation between these urinary pesticides levels and the severity of fatty liver disease (FLD).

METHODS

All participants' fat fraction (FF) values were recorded by MRI (Magnetic resonance imaging). First-morning urine samples were collected from 53 participants from Fuyang Peoples'Hospital. The levels of three urinary pesticides were measured using β-glucuronidase hydrolysis followed by a. The results were analyzed by using Pearson correlation analysis and binary logistic regression analysis to reveal the correlation between three urinary pesticides and the severity of fatty liver.

RESULTS

53 individuals were divided into 3 groups based on the results from MRI, with 20 cases in the normal control group, 16 cases in the mild fatty liver group, and 17 cases in the moderate and severe fatty liver group. Urinary chlorpyrifos level was increased along with the increase of the severity of fatty liver. Urinary paraquat level was significantly higher both in the low-grade fatty liver group and moderate & serve grade fatty liver group compared with the control group. No significant differences in urinary cyproconazole levels were observed among the three groups. Furthermore, urinary chlorpyrifos and paraquat levels were positively correlated with FF value. And chlorpyrifos was the risk factor that may be involved in the development of FLD and Receiver Operating Characteristic curve (ROC curve) analysis showed that chlorpyrifos and paraquat may serve as potential predictors of FLD.

CONCLUSION

The present findings indicate urinary chlorpyrifos and paraquat were positively correlated with the severity of fatty liver. Moreover, urinary chlorpyrifos and paraquat have the potential to be considered as the predictors for development of FLD. Thus, this study may provide a new perspective from the environmental factors for the diagnosis, prevention, and treatment of FLD.

A Comparative Study of Two Radiomics-Based Blood Flow Modes with Thyroid Imaging Reporting and Data System in Predicting Malignancy of Thyroid Nodules and Reducing Unnecessary Fine-Needle Aspiration Rate

RATIONALE AND OBJECTIVES

We aimed to compare superb microvascular imaging (SMI)-based radiomics methods, and contrast-enhanced ultrasound (CEUS)-based radiomics methods to the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) for classifying thyroid nodules (TNs) and reducing unnecessary fine-needle aspiration biopsy (FNAB) rate.

MATERIALS AND METHODS

This retrospective study enrolled a dataset of 472 pathologically confirmed TNs. Radiomics characteristics were extracted from B-mode ultrasound (BMUS), SMI, and CEUS images, respectively. After eliminating redundant features, four radiomics scores (Rad-scores) were constructed. Using multivariable logistic regression analysis, four radiomics prediction models incorporating Rad-score and corresponding US features were constructed and validated in terms of discrimination, calibration, decision curve analysis, and unnecessary FNAB rate.

RESULTS

The diagnostic performance of the BMUS + SMI radiomics method was better than ACR TI-RADS (area under the curve [AUC]: 0.875 vs. 0.689 for the training cohort, 0.879 vs. 0.728 for the validation cohort) (P < 0.05), and comparable with BMUS + CEUS radiomics method (AUC: 0.875 vs. 0.878 for the training cohort, 0.879 vs. 0.865 for the validation cohort) (P > 0.05). Decision curve analysis showed that the BMUS+SMI radiomics method could achieve higher net benefits than the BMUS radiomics method and ACR TI-RADS when the threshold probability was between 0.13 and 0.88 in the entire cohort. When applying the BMUS+SMI radiomics method, the unnecessary FNAB rate reduced from 43.4% to 13.9% in the training cohort and from 45.6% to 18.0% in the validation cohorts in comparison to ACR TI-RADS.

CONCLUSION

The dual-modal SMI-based radiomics method is convenient and economical and can be an alternative to the dual-modal CEUS-based radiomics method in helping radiologists select the optimal clinical strategy for TN management.

Study on diagnosis of thyroid nodules based on convolutional neural network

OBJECTIVE

An artificial intelligence (AI) algorithm based on convolutional neural networks was used in ultrasound diagnosis in order to evaluate its performance in judging the nature of thyroid nodules and nodule classification.

METHODS

A total of 105 patients with thyroid nodules confirmed by surgery or biopsy were retrospectively analyzed. The properties, characteristics, and classification of thyroid nodules were evaluated by sonographers and by AI to obtain combined diagnoses. Receiver operating characteristic curves were generated to evaluate the performance of AI, the sonographer, and their combined effort in diagnosing the nature of thyroid nodules and classifying their characteristics. In the diagnosis of thyroid nodules with solid components, hypoechoic appearance, indistinct borders, Anteroposterior/transverse diameter ratio > 1(A/T > 1), and calcification performed by sonographers and by AI, the properties exhibited statistically significant differences.

RESULTS

Sonographers had a sensitivity of 80.7%, specificity of 73.7%, accuracy of 79.0%, and area under the curve (AUC) of 0.751 in the diagnosis of benign and malignant thyroid nodules. AI had a sensitivity of 84.5%, specificity of 81.0%, accuracy of 84.7%, and AUC of 0.803. The combined AI and sonographer diagnosis had a sensitivity of 92.1%, specificity of 86.3%, accuracy of 91.7%, and AUC of 0.910.

CONCLUSION

The efficacy of a combined diagnosis for benign and malignant thyroid nodules is higher than that of an AI-based diagnosis alone or a sonographer-based diagnosis alone. The combined diagnosis can reduce unnecessary fine-needle aspiration biopsy procedures and better evaluate the necessity of surgery in clinical practice.

Integration of Artificial Intelligence Decision Aids to Reduce Workload and Enhance Efficiency in Thyroid Nodule Management

Importance

To optimize the integration of artificial intelligence (AI) decision aids and reduce workload in thyroid nodule management, it is critical to incorporate personalized AI into the decision-making processes of radiologists with varying levels of expertise.

Objective

To develop an optimized integration of AI decision aids for reducing radiologists' workload while maintaining diagnostic performance compared with traditional AI-assisted strategy.

Design, Setting, and Participants

In this diagnostic study, a retrospective set of 1754 ultrasonographic images of 1048 patients with 1754 thyroid nodules from July 1, 2018, to July 31, 2019, was used to build an optimized strategy based on how 16 junior and senior radiologists incorporated AI-assisted diagnosis results with different image features. In the prospective set of this diagnostic study, 300 ultrasonographic images of 268 patients with 300 thyroid nodules from May 1 to December 31, 2021, were used to compare the optimized strategy with the traditional all-AI strategy in terms of diagnostic performance and workload reduction. Data analyses were completed in September 2022.

Main Outcomes and Measures

The retrospective set of images was used to develop an optimized integration of AI decision aids for junior and senior radiologists based on the selection of AI-assisted significant or nonsignificant features. In the prospective set of images, the diagnostic performance, time-based cost, and assisted diagnosis were compared between the optimized strategy and the traditional all-AI strategy.

Results

The retrospective set included 1754 ultrasonographic images from 1048 patients (mean [SD] age, 42.1 [13.2] years; 749 women [71.5%]) with 1754 thyroid nodules (mean [SD] size, 16.4 [10.6] mm); 748 nodules (42.6%) were benign, and 1006 (57.4%) were malignant. The prospective set included 300 ultrasonographic images from 268 patients (mean [SD] age, 41.7 [14.1] years; 194 women [72.4%]) with 300 thyroid nodules (mean [SD] size, 17.2 [6.8] mm); 125 nodules (41.7%) were benign, and 175 (58.3%) were malignant. For junior radiologists, the ultrasonographic features that were not improved by AI assistance included cystic or almost completely cystic nodules, anechoic nodules, spongiform nodules, and nodules smaller than 5 mm, whereas for senior radiologists the features that were not improved by AI assistance were cystic or almost completely cystic nodules, anechoic nodules, spongiform nodules, very hypoechoic nodules, nodules taller than wide, lobulated or irregular nodules, and extrathyroidal extension. Compared with the traditional all-AI strategy, the optimized strategy was associated with increased mean task completion times for junior radiologists (reader 11, from 15.2 seconds [95% CI, 13.2-17.2 seconds] to 19.4 seconds [95% CI, 15.6-23.3 seconds]; reader 12, from 12.7 seconds [95% CI, 11.4-13.9 seconds] to 15.6 seconds [95% CI, 13.6-17.7 seconds]), but shorter times for senior radiologists (reader 14, from 19.4 seconds [95% CI, 18.1-20.7 seconds] to 16.8 seconds [95% CI, 15.3-18.3 seconds]; reader 16, from 12.5 seconds [95% CI, 12.1-12.9 seconds] to 10.0 seconds [95% CI, 9.5-10.5 seconds]). There was no significant difference in sensitivity (range, 91%-100%) or specificity (range, 94%-98%) between the 2 strategies for readers 11 to 16.

Conclusions and Relevance

This diagnostic study suggests that an optimized AI strategy in thyroid nodule management may reduce diagnostic time-based costs without sacrificing diagnostic accuracy for senior radiologists, while the traditional all-AI strategy may still be more beneficial for junior radiologists.

A Series-Based Deep Learning Approach to Lung Nodule Image Classification

Although many studies have shown that deep learning approaches yield better results than traditional methods based on manual features, CADs methods still have several limitations. These are due to the diversity in imaging modalities and clinical pathologies. This diversity creates difficulties because of variation and similarities between classes. In this context, the new approach from our study is a hybrid method that performs classifications using both medical image analysis and radial scanning series features. Hence, the areas of interest obtained from images are subjected to a radial scan, with their centers as poles, in order to obtain series. A U-shape convolutional neural network model is then used for the 4D data classification problem. We therefore present a novel approach to the classification of 4D data obtained from lung nodule images. With radial scanning, the eigenvalue of nodule images is captured, and a powerful classification is performed. According to our results, an accuracy of 92.84% was obtained and much more efficient classification scores resulted as compared to recent classifiers.

Endocrine Tumor Classification via Machine-Learning-Based Elastography: A Systematic Scoping Review

Elastography complements traditional medical imaging modalities by mapping tissue stiffness to identify tumors in the endocrine system, and machine learning models can further improve diagnostic accuracy and reliability. Our objective in this review was to summarize the applications and performance of machine-learning-based elastography on the classification of endocrine tumors. Two authors independently searched electronic databases, including PubMed, Scopus, Web of Science, IEEEXpress, CINAHL, and EMBASE. Eleven (n = 11) articles were eligible for the review, of which eight (n = 8) focused on thyroid tumors and three (n = 3) considered pancreatic tumors. In all thyroid studies, the researchers used shear-wave ultrasound elastography, whereas the pancreas researchers applied strain elastography with endoscopy. Traditional machine learning approaches or the deep feature extractors were used to extract the predetermined features, followed by classifiers. The applied deep learning approaches included the convolutional neural network (CNN) and multilayer perceptron (MLP). Some researchers considered the mixed or sequential training of B-mode and elastographic ultrasound data or fusing data from different image segmentation techniques in machine learning models. All reviewed methods achieved an accuracy of ≥80%, but only three were ≥90% accurate. The most accurate thyroid classification (94.70%) was achieved by applying sequential training CNN; the most accurate pancreas classification (98.26%) was achieved using a CNN-long short-term memory (LSTM) model integrating elastography with B-mode and Doppler images.

A comparison of artificial intelligence versus radiologists in the diagnosis of thyroid nodules using ultrasonography: a systematic review and meta-analysis

BACKGROUND

Thyroid nodules are common. Ultrasonography (US) is the first investigation for thyroid nodules. Artificial Intelligence (AI) is widely integrated into medical diagnosis to provide additional information. The primary objective of this study was to accumulate the pooled sensitivity and specificity between all available AI and radiologists using thyroid US imaging. The secondary objective was to compare AI's diagnostic performance to that of radiologists.

MATERIALS AND METHODS

A systematic review meta-analysis. PubMed, Scopus, Web of Science, and Cochrane Library data were searched for studies from inception until June 11, 2020.

RESULTS

Twenty five studies were included in this meta-analysis. The pooled sensitivity and specificity of AI were 0.86 (95% CI 0.81-0.91) and 0.78 (95% CI 0.73-0.83), respectively. The pooled sensitivity and specificity of radiologists were 0.85 (95% CI 0.80-0.89) and 0.82 (95% CI 0.77-0.86), respectively. The accuracy of AI and radiologists is equivalent in terms of AUC [AI 0.89 (95% CI 0.86-0.92), radiologist 0.91 (95% CI 0.88-0.93)]. The diagnostic odd ratio (DOR) between AI 23.10 (95% CI 14.20-37.58) and radiologists 27.12 (95% CI 17.45-42.16) had no statistically significant difference (P = 0.56). Meta-regression analysis revealed that Deep Learning AI had significantly greater sensitivity and specificity than classic machine learning AI (P < 0.001).

CONCLUSION

AI demonstrated comparable performance to radiologists in diagnosing benign and malignant thyroid nodules using ultrasonography. Additional research to establish its equivalency should be conducted.

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.

Performance of novel deep learning network with the incorporation of the automatic segmentation network for diagnosis of breast cancer in automated breast ultrasound

OBJECTIVE

To develop novel deep learning network (DLN) with the incorporation of the automatic segmentation network (ASN) for morphological analysis and determined the performance for diagnosis breast cancer in automated breast ultrasound (ABUS).

METHODS

A total of 769 breast tumors were enrolled in this study and were randomly divided into training set and test set at 600 vs. 169. The novel DLNs (Resent v2, ResNet50 v2, ResNet101 v2) added a new ASN to the traditional ResNet networks and extracted morphological information of breast tumors. The accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), area under the receiver operating characteristic (ROC) curve (AUC), and average precision (AP) were calculated. The diagnostic performances of novel DLNs were compared with those of two radiologists with different experience.

RESULTS

The ResNet34 v2 model had higher specificity (76.81%) and PPV (82.22%) than the other two, the ResNet50 v2 model had higher accuracy (78.11%) and NPV (72.86%), and the ResNet101 v2 model had higher sensitivity (85.00%). According to the AUCs and APs, the novel ResNet101 v2 model produced the best result (AUC 0.85 and AP 0.90) compared with the remaining five DLNs. Compared with the novice radiologist, the novel DLNs performed better. The F1 score was increased from 0.77 to 0.78, 0.81, and 0.82 by three novel DLNs. However, their diagnostic performance was worse than that of the experienced radiologist.

CONCLUSIONS

The novel DLNs performed better than traditional DLNs and may be helpful for novice radiologists to improve their diagnostic performance of breast cancer in ABUS.

KEY POINTS

• A novel automatic segmentation network to extract morphological information was successfully developed and implemented with ResNet deep learning networks. • The novel deep learning networks in our research performed better than the traditional deep learning networks in the diagnosis of breast cancer using ABUS images. • The novel deep learning networks in our research may be useful for novice radiologists to improve diagnostic performance.

Lack of Association Between Helicobacter pylori Infection and the Risk of Thyroid Nodule Types: A Multicenter Case-Control Studyin China

The prevalence of Helicobacter pylori infection is high worldwide, while numerous research has focused on unraveling the relationship between H. pylori infection and extragastric diseases. Although H. pylori infection has been associated with thyroid diseases, including thyroid nodule (TN), the relationship has mainly focused on potential physiological mechanisms and has not been validated by large population epidemiological investigations. Therefore, we thus designed a case-control study comprising participants who received regular health examination between 2017 and 2019. The cases and controls were diagnosed via ultrasound, while TN types were classified according to the guidelines of the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS). Moreover, H. pylori infection was determined by C14 urea breath test, while its relationship with TN type risk and severity was analyzed using binary and ordinal logistic regression analyses. A total of 43,411 participants, including 13,036 TN patients and 30,375 controls, were finally recruited in the study. The crude odds ratio (OR) was 1.07 in Model 1 (95% CI = 1.03-1.14) without adjustment compared to the H. pylori non-infection group. However, it was negative in Model 2 (OR = 1.02, 95% CI = 0.97-1.06) after being adjusted for gender, age, body mass index (BMI), and blood pressure and in Model 3 (OR = 1.01, 95% CI = 0.97-1.06) after being adjusted for total cholesterol, triglyceride, low-density lipoprotein, and high-density lipoprotein on the basis of Model 2. Control variables, including gender, age, BMI, and diastolic pressure, were significantly correlated with the risk of TN types. Additionally, ordinal logistic regression results revealed that H. pylori infection was positively correlated with malignant differentiation of TN (Model 1: OR = 1.06, 95% CI = 1.02-1.11), while Model 2 and Model 3 showed negative results (Model 2: OR = 1.01, 95% CI = 0.96-1.06; Model 3: OR = 1.01, 95% CI = 0.96-1.05). In conclusion, H. pylori infection was not significantly associated with both TN type risk and severity of its malignant differentiation. These findings provide relevant insights for correcting possible misconceptions regarding TN type pathogenesis and will help guide optimization of therapeutic strategies for thyroid diseases.

Modified Models for Predicting Malignancy Using Ultrasound Characters Have High Accuracy in Thyroid Nodules With Small Size

To assess the malignancy risk of thyroid nodules, ten ultrasound characteristics are suggested as key diagnostic markers. The European Thyroid Association Guidelines (EU-TIRADS) and 2015 American Thyroid Association Management Guidelines (2015ATA) are mainly used for ultrasound malignancy risk stratification, but both are less accurate and do not appropriatetly classify high risk patients in clinical examination. Previous studies focus on papillary thyroid carcinoma (PTC), but follicular thyroid carcinoma (FTC) and medullary thyroid carcinoma (MTC) remained to be characterized. Thus, this study aimed to determine the diagnostic accuracy and establish models using all ultrasound features including the nodule size for predicting the malignancy of thyroid nodules (PTC, FTC, and MTC) in China. We applied logistic regression to the data of 1,500 patients who received medical treatment in Shanghai and Fujian. Ultrasound features including taller-than-wide shape and invasion of the thyroid capsule showed high odds ratio (OR 19.329 and 4.672) for PTC in this dataset. Invasion of the thyroid also showed the highest odds ratio (OR = 8.10) for MTC. For FTC, the halo sign has the highest odds ratio (OR = 13.40). Four ultrasound features revealed distinct OR in PTC nodule groups with different sizes. In this study, we constructed a logistic model with accuracy up to 80%. In addition, this model revealed more accuracy than TIRADS in 4b and 4c category nodules. Hence, this model could well predict malignancy in small nodules and classify high-risk patients.

The impact of the use of the ACR-TIRADS as a screening tool for thyroid nodules in a cancer center

BACKGROUND

The Thyroid Imaging Reporting and Data System (TIRADS) was created to assess risk of thyroid nodules through ultrasound. Plenty classifications methods for thyroid nodules have already been created, but none of them have yet achieved global utilization. This study analyzed the performance of the American College of Radiology (ACR) TIRADS, its reproducibility and the impact of its utilization as a screening method in a large Cancer Center cohort.

METHODS

Thyroid nodules which underwent fine-needle aspiration (FNA) in a 1-year period were selected, with their ultrasound images retrospectively classified according to the ACR TI-RADS. Cytological evaluation of the nodules and final histology (whenever available) was used to assess risk of neoplasm (RON) and risk of malignancy (ROM) associated to each ACR-TIRADS category. Further analyses were also carried out according to recommendation or not of FNA by the ACR-TIRADS and nodule size. Inter-observer agreement for the system was also assessed.

RESULTS

A total of 1112 thyroid nodules were included. RON for each category according to final cytological diagnosis was 0% for TR1 and TR2, 2.1% for TR3; 15.6% for TR4 and 68.9% for TR5. No significant difference was observed between the RON of the categories for cases above or below 1.0 cm. Nodules that met the criteria for FNA had 3 times greater chance of a positive outcome. Substantial agreement (kappa 0.77) was seen between two different observers.

CONCLUSIONS

ACR TI-RADS scoring system has demonstrated to be an accurate method to stratify thyroid nodules in a Cancer Center, with a high reproducibility.

A Comparative Analysis of Two Machine Learning-Based Diagnostic Patterns with Thyroid Imaging Reporting and Data System for Thyroid Nodules: Diagnostic Performance and Unnecessary Biopsy Rate

Background: The risk stratification system of the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) for thyroid nodules is affected by low diagnostic specificity. Machine learning (ML) methods can optimize the diagnostic performance in medical image analysis. However, it is unknown which ML-based diagnostic pattern is more effective in improving diagnostic performance for thyroid nodules and reducing nodule biopsies. Therefore, we compared ML-assisted visual approaches and radiomics approaches with ACR TI-RADS in diagnostic performance and unnecessary fine-needle aspiration biopsy (FNAB) rate for thyroid nodules. Methods: This retrospective study evaluated a data set of ultrasound (US) and shear wave elastography (SWE) images in patients with biopsy-proven thyroid nodules (≥1 cm) from the Shanghai Tenth People's Hospital (743 nodules in 720 patients from September 2017 to January 2019) and an independent test data set from the Ma'anshan People's Hospital (106 nodules in 102 patients from February 2019 to April 2019). Six US features and five SWE parameters from the radiologists' interpretation were used for building the ML-assisted visual approaches. The radiomics features extracted from the US and SWE images were used with ML methods for developing the radiomics approaches. The diagnostic performance for differentiating thyroid nodules and the unnecessary FNAB rate of the ML-assisted visual approaches and the radiomics approaches were compared with ACR TI-RADS. Results: The ML-assisted US visual approach had the best diagnostic performance than the US radiomics approach and ACR TI-RADS (area under the curve [AUC]: 0.900 vs. 0.789 vs. 0.689 for the validation data set, 0.917 vs. 0.770 vs. 0.681 for the test data set). After adding SWE, the ML-assisted visual approach had a better diagnostic performance than US alone (AUC: 0.951 vs. 0.900 for the validation data set, 0.953 vs. 0.917 for the test data set). When applying the ML-assisted US+SWE visual approach, the unnecessary FNAB rate decreased from 30.0% to 4.5% in the validation data set and from 37.7% to 4.7% in the test data set in comparison to ACR TI-RADS. Conclusions: The ML-assisted dual modalities visual approach can assist radiologists to diagnose thyroid nodules more effectively and considerably reduce the unnecessary FNAB rate in the clinical management of thyroid nodules.

Artificial Intelligence for Personalized Medicine in Thyroid Cancer: Current Status and Future Perspectives

Thyroid cancers (TC) have increasingly been detected following advances in diagnostic methods. Risk stratification guided by refined information becomes a crucial step toward the goal of personalized medicine. The diagnosis of TC mainly relies on imaging analysis, but visual examination may not reveal much information and not enable comprehensive analysis. Artificial intelligence (AI) is a technology used to extract and quantify key image information by simulating complex human functions. This latent, precise information contributes to stratify TC on the distinct risk and drives tailored management to transit from the surface (population-based) to a point (individual-based). In this review, we started with several challenges regarding personalized care in TC, for example, inconsistent rating ability of ultrasound physicians, uncertainty in cytopathological diagnosis, difficulty in discriminating follicular neoplasms, and inaccurate prognostication. We then analyzed and summarized the advances of AI to extract and analyze morphological, textural, and molecular features to reveal the ground truth of TC. Consequently, their combination with AI technology will make individual medical strategies possible.

Visual Interpretability in Computer-Assisted Diagnosis of Thyroid Nodules Using Ultrasound Images

BACKGROUND The number of studies on deep learning in artificial intelligence (AI)-assisted diagnosis of thyroid nodules is increasing. However, it is difficult to explain what the models actually learn in artificial intelligence-assisted medical research. Our aim is to investigate the visual interpretability of the computer-assisted diagnosis of malignant and benign thyroid nodules using ultrasound images. MATERIAL AND METHODS We designed and implemented 2 experiments to test whether our proposed model learned to interpret the ultrasound features used by ultrasound experts to diagnose thyroid nodules. First, in an anteroposterior/transverse (A/T) ratio experiment, multiple models were trained by changing the A/T ratio of the original nodules, and their classification, accuracy, sensitivity, and specificity were tested. Second, in a visualization experiment, class activation mapping used global average pooling and a fully connected layer to visualize the neural network to show the most important features. We also examined the importance of data preprocessing. RESULTS The A/T ratio experiment showed that after changing the A/T ratio of the nodules, the accuracy of the neural network model was reduced by 9.24-30.45%, indicating that our neural network model learned the A/T ratio information of the nodules. The visual experiment results showed that the nodule margins had a strong influence on the prediction of the neural network. CONCLUSIONS This study was an active exploration of interpretability in the deep learning classification of thyroid nodules. It demonstrated the neural network-visualized model focused on irregular nodule margins and the A/T ratio to classify thyroid nodules.

S-Detect Software vs. EU-TIRADS Classification: A Dual-Center Validation of Diagnostic Performance in Differentiation of Thyroid Nodules

Computer-aided diagnosis (CAD) and other risk stratification systems may improve ultrasound image interpretation. This prospective study aimed to compare the diagnostic performance of CAD and the European Thyroid Imaging Reporting and Data System (EU-TIRADS) classification applied by physicians with S-Detect 2 software CAD based on Korean Thyroid Imaging Reporting and Data System (K-TIRADS) and combinations of both methods (MODELs 1 to 5). In all, 133 nodules from 88 patients referred to thyroidectomy with available histopathology or with unambiguous results of cytology were included. The S-Detect system, EU-TIRADS, and mixed MODELs 1–5 for the diagnosis of thyroid cancer showed a sensitivity of 89.4%, 90.9%, 84.9%, 95.5%, 93.9%, 78.9% and 93.9%; a specificity of 80.6%, 61.2%, 88.1%, 53.7%, 73.1%, 89.6% and 80.6%; a positive predictive value of 81.9%, 69.8%, 87.5%, 67%, 77.5%, 88.1% and 82.7%; a negative predictive value of 88.5%, 87.2%, 85.5%, 92.3%, 92.5%, 81.1% and 93.1%; and an accuracy of 85%, 75.9%, 86.5%, 74.4%, 83.5%, 84.2%, and 87.2%, respectively. Comparison showed superiority of the similar MODELs 1 and 5 over other mixed models as well as EU-TIRADS and S-Detect used alone (p-value < 0.05). S-Detect software is characterized with high sensitivity and good specificity, whereas EU-TIRADS has high sensitivity, but rather low specificity. The best diagnostic performance in malignant thyroid nodule (TN) risk stratification was obtained for the combined model of S-Detect (“possibly malignant” nodule) and simultaneously obtaining 4 or 5 points (MODEL 1) or exactly 5 points (MODEL 5) on the EU-TIRADS scale.