Thyroid Nodules In Ultrasound Images Research Articles

An Improved Object Detection Algorithm for Thyroid Nodule Ultrasound Image Based on Faster R-CNN

To propose an improved algorithm for thyroid nodule object detection based on Faster R-CNN so as to improve the detection precision of thyroid nodules in ultrasound images. The algorithm used ResNeSt50 combined with deformable convolution (DC) as the backbone network to improve the detection effect of irregularly shaped nodules. Feature pyramid networks (FPN) and Region of Interest (RoI) Align were introduced in the back of the trunk network. The former was used to reduce missed or mistaken detection of thyroid nodules, and the latter was used to improve the detection precision of small nodules. To improve the generalization ability of the model, parameters were updated during backpropagation with an optimizer improved by Sharpness-Aware Minimization (SAM). In this experiment, 6 261 thyroid ultrasound images from the Affiliated Hospital of Xuzhou Medical University and the First Hospital of Nanjing were used to compare and evaluate the effectiveness of the improved algorithm. According to the findings, the algorithm showed optimization effect to a certain degree, with the AP50 of the final test set being as high as 97.4% and AP@50:5:95 also showing a 10.0% improvement compared with the original model. Compared with both the original model and the existing models, the improved algorithm had higher detection precision and improved capacity to detect thyroid nodules with better accuracy and precision. In particular, the improved algorithm had a higher recall rate under the requirement of lower detection frame precision. The improved method proposed in the study is an effective object detection algorithm for thyroid nodules and can be used to detect thyroid nodules with accuracy and precision.

SK-Unet++: An improved Unet++ network with adaptive receptive fields for automatic segmentation of ultrasound thyroid nodule images.

The quality of segmentation of thyroid nodules in ultrasound images is a crucial factor in preventing the cancerization of thyroid nodules. However, the existing standards for the ultrasound imaging of cancerous nodules have limitations, and changes of the echo pattern of thyroid nodules pose challenges in accurately segmenting nodules, which can affect the diagnostic results of medical professionals. The aim of this study is to address the challenges related to segmentation accuracy due to noise, low contrast, morphological scale variations, and blurred edges of thyroid nodules in ultrasound images and improve the accuracy of ultrasound-based thyroid nodule segmentation, thereby aiding the clinical diagnosis of thyroid nodules. In this study, the dataset of thyroid ultrasound images was obtained from Hunan Provincial People's Hospital, consisting of a total of 3572 samples used for the training, validation, and testing of this model at a ratio of 8:1:1. A novel SK-Unet++ network was used to enhance the segmentation accuracy of thyroid nodules. SK-Unet++ is a novel deep learning architecture that adds the adaptive receptive fields based on the selective kernel (SK) attention mechanisms into the Unet++ network. The convolution blocks of the original UNet++ encoder part were replaced with finer SK convolution blocks in SK-Unet++. First, multiple skip connections were incorporated so that SK-Unet++ can make information from previous layers of the neural network to bypass certain layers and directly propagate to subsequent layers. The feature maps of the corresponding locations were fused on the channel, resulting in enhanced segmentation accuracy. Second, we added the adaptive receptive fields. The adaptive receptive fields were used to capture multiscale spatial features better by dynamically adjusting its receptive field. The assessment metrics contained dice similarity coefficient (Dsc), accuracy (Acc), precision (Pre), recall (Re), and Hausdorff distance, and all comparison experiments used the paired t-tests to assess whether statistically significant performance differences existed (p<0.05). And to address the multi-comparison problem, we performed the false discovery rate (FDR) correction after the test. The segmentation model had an Acc of 80.6%, Dsc of 84.7%, Pre of 77.5%, Re of 71.7%, and an average Hausdorff distance of 15.80mm. Ablation experimental results demonstrated that each module in the network could contribute to the improved performance (p<0.05) and determined the best combination of parameters. A comparison with other state-of-the-art methods showed that SK-Unet++ significantly outperformed them in terms of segmentation performance (p<0.05), with a more accurate segmentation contour. Additionally, the adaptive weight changes of the SK module were monitored during the training process, and the resulting change curves demonstrated their convergence. Our proposed method demonstrates favorable performance in the segmentation of ultrasound images of thyroid nodules. Results confirmed that SK-Unet++ is a feasible and effective method for the automatic segmentation of thyroid nodules in ultrasound images. The high accuracy achieved by our method can facilitate efficient screening of patients with thyroid nodules, ultimately reducing the workload of clinicians and radiologists.

DPAM-PSPNet: ultrasonic image segmentation of thyroid nodule based on dual-path attention mechanism

Objective. Deep learning has demonstrated its versatility in the medical field, particularly in medical image segmentation, image classification, and other forms of automated diagnostics. The clinical diagnosis of thyroid nodules requires radiologists to locate nodules, diagnose conditions based on nodule boundaries, textures and their experience. This task is labor-intensive and tiring; therefore, an automated system for accurate thyroid nodule segmentation is essential. In this study, a model named DPAM-PSPNet was proposed, which automatically segments nodules in thyroid ultrasound images and enables to segment malignant nodules precisely. Approach. In this paper, accurate segmentation of nodule edges is achieved by introducing the dual path attention mechanism (DPAM) in PSPNet. In one channel, it captures global information with a lightweight cross-channel interaction mechanism. In other channel, it focus on nodal margins and surrounding information through the residual bridge network. We also updated the integrated loss function to accommodate the DPAM-PSPNet. Main results. The DPAM-PSPNet was tested against the classical segmentation model. Ablation experiments were designed for the two-path attention mechanism and the new loss function, and generalization experiments were designed on the public dataset. Our experimental results demonstrate that DPAM-PSPNet outperforms other existing methods in various evaluation metrics. In the model comparison experiments, it achieved performance with an mIOU of 0.8675, mPA of 0.9357, mPrecision of 0.9202, and Dice coefficient of 0.9213. Significance. The DPAM-PSPNet model can segment thyroid nodules in ultrasound images with little training data and generate accurate boundary regions for these nodules.

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

Based on ultrasound (US) images, this study aimed to detect and quantify calcifications of thyroid nodules, which are regarded as one of the most important features in US diagnosis of thyroid cancer, and to further investigate the value of US calcifications in predicting the risk of lymph node metastasis (LNM) in papillary thyroid cancer (PTC). Based on the DeepLabv3+ networks, 2992 thyroid nodules in US images were used to train a model to detect thyroid nodules, of which 998 were used to train a model to detect and quantify calcifications. A total of 225 and 146 thyroid nodules obtained from two centers, respectively, were used to test the performance of these models. A logistic regression method was used to construct the predictive models for LNM in PTCs. Calcifications detected by the network model and experienced radiologists had an agreement degree of above 90%. The novel quantitative parameters of US calcification defined in this study showed a significant difference between PTC patients with and without cervical LNM (p < 0.05). The calcification parameters were beneficial to predicting the LNM risk in PTC patients. The LNM prediction model using these calcification parameters combined with patient age and other US nodular features showed a higher specificity and accuracy than the calcification parameters alone. Our models not only detect the calcifications automatically, but also have value in predicting cervical LNM risk of PTC patients, thereby making it possible to investigate the relationship between calcifications and highly invasive PTC in detail. Due to the high association of US microcalcifications with thyroid cancers, our model will contribute to the differential diagnosis of thyroid nodules in daily practice. • We developed an ML-based network model for automatically detecting and quantifying calcifications within thyroid nodules in US images. • Three novel parameters for quantifying US calcifications were defined and verified. • These US calcification parameters showed value in predicting the risk of cervical LNM in PTC patients.

FBN: Weakly Supervised Thyroid Nodule Segmentation Optimized by Online Foreground and Background

The main objective of the work described here was to train a semantic segmentation model using classification data for thyroid nodule ultrasound images to reduce the pressure of obtaining pixel-level labeled data sets. Furthermore, we improved the segmentation performance of the model by mining the image information to narrow the gap between weakly supervised semantic segmentation (WSSS) and fully supervised semantic segmentation. Most WSSS methods use a class activation map (CAM) to generate segmentation results. However, the lack of supervision information makes it difficult for a CAM to highlight the object region completely. Therefore, we here propose a novel foreground and background pair (FB-Pair) representation method, which consists of high- and low-response regions highlighted by the original CAM-generated online in the original image. During training, the original CAM is revised using the CAM generated by the FB-Pair. In addition, we design a self-supervised learning pretext task based on FB-Pair, which requires the model to predict whether the pixels in FB-Pair are from the original image during training. After this task, the model will accurately distinguish between different categories of objects. Experiments on the thyroid nodule ultrasound image (TUI) data set revealed that our proposed method outperformed existing methods, with a 5.7% improvement in the mean intersection-over-union (mIoU) performance of segmentation compared with the second-best method and a reduction to 2.9% in the difference between the performance of benign and malignant nodules. Our method trains a well-performing segmentation model on ultrasound images of thyroid nodules using only classification data. In addition, we determined that CAM can take full advantage of the information in the images to highlight the target regions more accurately and thus improve the segmentation performance.

Multistep Automated Data Labelling Procedure (MADLaP) for thyroid nodules on ultrasound: An artificial intelligence approach for automating image annotation

Machine learning (ML) for diagnosis of thyroid nodules on ultrasound is an active area of research. However, ML tools require large, well-labeled datasets, the curation of which is time-consuming and labor-intensive. The purpose of our study was to develop and test a deep-learning-based tool to facilitate and automate the data annotation process for thyroid nodules; we named our tool Multistep Automated Data Labelling Procedure (MADLaP). MADLaP was designed to take multiple inputs including pathology reports, ultrasound images, and radiology reports. Using multiple step-wise ‘modules’ including rule-based natural language processing, deep-learning-based imaging segmentation, and optical character recognition, MADLaP automatically identified images of a specific thyroid nodule and correctly assigned a pathology label. The model was developed using a training set of 378 patients across our health system and tested on a separate set of 93 patients. Ground truths for both sets were selected by an experienced radiologist. Performance metrics including yield (how many labeled images the model produced) and accuracy (percentage correct) were measured using the test set. MADLaP achieved a yield of 63 % and an accuracy of 83 %. The yield progressively increased as the input data moved through each module, while accuracy peaked part way through. Error analysis showed that inputs from certain examination sites had lower accuracy (40 %) than the other sites (90 %, 100 %). MADLaP successfully created curated datasets of labeled ultrasound images of thyroid nodules. While accurate, the relatively suboptimal yield of MADLaP exposed some challenges when trying to automatically label radiology images from heterogeneous sources. The complex task of image curation and annotation could be automated, allowing for enrichment of larger datasets for use in machine learning development.

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images.

In recent years, the incidence of thyroid cancer has been increasing. Thyroid nodule detection is critical for both the detection and treatment of thyroid cancer. Convolutional neural networks (CNNs) have achieved good results in thyroid ultrasound image analysis tasks. However, due to the limited valid receptive field of convolutional layers, CNNs fail to capture long-range contextual dependencies, which are important for identifying thyroid nodules in ultrasound images. Transformer networks are effective in capturing long-range contextual information. Inspired by this, we propose a novel thyroid nodule detection method that combines the Swin Transformer backbone and Faster R-CNN. Specifically, an ultrasound image is first projected into a 1D sequence of embeddings, which are then fed into a hierarchical Swin Transformer. The Swin Transformer backbone extracts features at five different scales by utilizing shifted windows for the computation of self-attention. Subsequently, a feature pyramid network (FPN) is used to fuse the features from different scales. Finally, a detection head is used to predict bounding boxes and the corresponding confidence scores. Data collected from 2,680 patients were used to conduct the experiments, and the results showed that this method achieved the best mAP score of 44.8%, outperforming CNN-based baselines. In addition, we gained better sensitivity (90.5%) than the competitors. This indicates that context modeling in this model is effective for thyroid nodule detection.

Ultrasound Image Segmentation and Classification of Benign and Malignant Thyroid Nodules on the Basis of Deep Learning

This study aimed to investigate the effect of an image denoising algorithm based on weighted low-rank matrix restoration on thyroid nodule ultrasound images. A total of 1000 original ultrasound image data sets of thyroid nodules were selected as the study samples. The nodule segmentation data set of thyroid ultrasound region of interest (ROI) images was drawn and acquired. By introducing multiscale features and an attention mechanism to optimize the U-Net model, an ultrasound image segmentation model (F-U-Net) was constructed. The performance of the traditional U network model and full convolutional neural network model (FCN) was analyzed and compared by simulation experiments. The results showed that the dice coefficient, accuracy, and recall of the improved loss function in this study were significantly higher than those of the traditional cross entropy loss function and dice coefficient loss function, and the differences were statistically significant (P &lt; 0.05). The Dice coefficient, accuracy, and recall of the F-U-net model were significantly higher than those of the traditional FCN model and U-net model (P &lt; 0.05). The diagnostic sensitivity, specificity, accuracy, and positive predictive value of the F-U-net model for benign and malignant thyroid nodules were significantly higher than those of the FCN model and U-net model (P &lt; 0.05). In summary, the proposed F-U network can effectively process the ultrasound images of thyroid nodules, improve the image quality, and help to improve the diagnostic effect of benign and malignant thyroid nodules. It provides a data reference for segmentation and reconstruction of benign and malignant ultrasound images of thyroid nodules.

Fusing enhanced Transformer and large kernel CNN for malignant thyroid nodule segmentation

Quick and accurate diagnosis for malignant thyroid nodules via ultrasonography is a valuable but challenging task even for experienced radiologists because of the complexity and variability of ultrasound images. Many computer-aided diagnosis (CAD) methods have been proposed to assist radiologists by providing objective suggestions. However, most existing segmentation approaches lack the ability to keep precise shape information and capture global long-range dependencies. To settle the above issues, we propose a deep learning-based CAD method called Transformer fusing CNN Network (TCNet) to segment malignant thyroid nodules automatically. Our proposed TCNet contains a large kernel CNN branch and an enhanced Transformer branch. In the former branch, we devise a Large Kernel Module (LKM) to extract the precise shape features of malignant thyroid nodules in ultrasound images. While in the latter branch, we design an Enhanced Transformer Module (ETM) to establish the remote connection between thyroid nodule pixels. To integrate multiscale feature maps produced from different phases of both branches, we develop a Multiscale Fusion Module (MFM) to connect the two branches. We compare the proposed model with several current commonly used segmentation methods on the MTNS dataset and other public datasets. The experimental results demonstrate the superiority and effectiveness of our method.

Improving GAN Learning Dynamics for Thyroid Nodule Segmentation.

Thyroid nodules are lesions requiring diagnosis and follow-up. Tools for detecting and segmenting nodules can help physicians with this diagnosis. Besides immediate diagnosis, automated tools can also enable tracking of the probability of malignancy over time. This paper demonstrates a new algorithm for segmenting thyroid nodules in ultrasound images. The algorithm combines traditional supervised semantic segmentation with unsupervised learning using GANs. The hybrid approach has the potential to upgrade the semantic segmentation model's performance, but GANs have the well-known problems of unstable learning and mode collapse. To stabilize the training of the GAN model, we introduce the concept of closed-loop control of the gain on the loss output of the discriminator. We find gain control leads to smoother generator training and avoids the mode collapse that typically occurs when the discriminator learns too quickly relative to the generator. We also find that the combination of the supervised and unsupervised learning styles encourages both low-level accuracy and high-level consistency. As a test of the concept of controlled hybrid supervised and unsupervised semantic segmentation, we introduce a new model named the StableSeg GAN. The model uses DeeplabV3+ as the generator, Resnet18 as the discriminator, and uses PID control to stabilize the GAN learning process. The performance of the new model in terms of IoU is better than DeeplabV3+, with mean IoU of 81.26% on a challenging test set. The results of our thyroid nodule segmentation experiments show that StableSeg GANs have flexibility to segment nodules more accurately than either comparable supervised segmentation models or uncontrolled GANs.

Reducing Operation Costs of Thyroid Nodules Using Machine Learning Algorithms with Thyroid Nodules Scoring Systems

Continuous advancement in the health sector is essential to reduce costs and increase efficiency and quality of service. The widespread use of ultrasonography (USG) has made it possible to detect thyroid nodules with higher success rates. Some standard scoring systems have been developed to score thyroid nodules. Thyroid scoring systems are classification systems that determine the risk of cancer in thyroid nodules according to ultrasonographic characteristics and nodule size. Different scoring results for the same thyroid nodule may occur according to these different scoring systems, which can cause some unnecessary surgical interventions. In this study, some intelligent models are developed to assist thyroid scoring systems, with the aim to determine the correct surgical intervention and reduce operation costs by preventing unnecessary interventions and surgical procedures. The integration of current thyroid scoring systems (K-TIRADS, ACR-TIRADS, EU-TIRADS, ATA, and BTA) and machine learning methods provides radiologists and clinicians a decision-support mechanism in the evaluation of thyroid nodules. Correct diagnosis will help to reduce costs by helping prevent unnecessary procedures. The present dataset was retrospectively constructed using ultrasound images of thyroid nodules between 2014 and 2018. In determining the treatment process of thyroid nodules, Random Forest, Adaboost, J48 Decision Tree (J48 DT), and Support Vector Machine (SVM) models are used for increased prediction accuracy of thyroid scoring systems. The goal is to decrease redundant Fine Needle Aspiration (FNA) biopsies and surgical interventions of suspicious thyroid nodules. As a result of the study, higher degrees of accuracy are achieved in the determination of correct or incorrect surgical interventions of thyroid nodules using the J48 DT algorithm with the EU-TIRADS scoring system, with an accuracy rate of 99.7853%, compared to other classifiers.

A Deep Learning-Based Cascade Automatic Classification System for Malignant Thyroid Nodule Recognition in Ultrasound Image

As the most common malignancy in the endocrine system, thyroid cancer is often determined based on ultrasound images by professional radiologists based on ultrasound images. In this study, a cascade automatic classification system (CACS) is proposed to accurately achieve automatic segmentation and classification of thyroid nodules, reducing over-reliance on experienced doctors in the diagnosis. Firstly, the Res-UNet is used to segment the region of interest (ROI) of thyroid nodules in ultrasound images. Then, to overcome the limitations of the small sample size of medical training data as well as category imbalance, the modified Conditional Variational Auto-Encoder (CVAE), learning the probability distribution of the ultrasound images, is trained to produce new medical ultrasound images for ROI data augmentation. Finally, we perform the classification task by a three-branch ensemble network combined with clinical diagnostic features of thyroid nodule such as the aspect ratio, internal and external echoes, and edges. The proposed CACS can identify benign and malignant thyroid nodules with an accuracy of 0.892 and an AUC of 0.887, according to testing on real patient data.

Automatic classification of thyroid nodules in ultrasound images using a multi-task attention network guided by clinical knowledge

Thyroid cancer has been the most prevalent cancer in the recent three decades. Ultrasonography is one of the mainly used methods for diagnosing thyroid nodules. Several computer-aided diagnostic methods were proposed to aid radiologists in analyzing ultrasound images of the thyroid gland. Most methods, however, only determine the benignity or malignancy of the thyroid nodule and do not explain the decision-making process of them, which cannot gain the trustworthiness of clinicians because they are not consistent with the physician’s diagnostic process. In our work, we design a multi-task branching attention network in which each of the descriptors of the ACR TI-RADS lexicon is first classified. All respective scores are calculated to get the risk stratification of the nodule. Ultimately, based on the risk stratification, the benignity or malignancy of the nodule is determined. This work provides an automated method that incorporates the ACR TI-RADS characterization of thyroid nodules for detecting the level of risk and the benignity or malignancy of thyroid nodules. Thus the work establishes the trustworthiness of clinicians in deep learning models and improves physician efficiency and diagnostic rates to some extent compared to previous studies. For the diagnosis of thyroid nodules, evaluation indices including accuracy, sensitivity, and specificity were 93.55%, 93.86%, and 93.14%, respectively. The experiments show that our approach obtains comparable performance to most advanced methods in diagnosing ultrasound images of the thyroid nodules and is supported by explanations in clinical terms using the ACR TI-RADS lexicon.

Thyroid Nodule Segmentation in Ultrasound Image Based on Information Fusion of Suggestion and Enhancement Networks

Computer-aided diagnosis/detection (CADx) systems have been used to help doctors in improving the quality of diagnosis and treatment processes in many serious diseases such as breast cancer, brain stroke, lung cancer, and bone fracture. However, the performance of such systems has not been completely accurate. The key factor in CADx systems is to localize positive disease lesions from the captured medical images. This step is important as it is used not only to localize lesions but also to reduce the effect of noise and normal regions on the overall CADx system. In this research, we proposed a method to enhance the segmentation performance of thyroid nodules in ultrasound images based on information fusion of suggestion and enhancement segmentation networks. Experimental results with two open databases of thyroid digital image databases and 3DThyroid databases showed that our method resulted in a higher performance compared to current up-to-date methods.

Multi-Attribute Attention Network for Interpretable Diagnosis of Thyroid Nodules in Ultrasound Images.

Ultrasound (US) is the primary imaging technique for the diagnosis of thyroid cancer. However, accurate identification of nodule malignancy is a challenging task that can elude less-experienced clinicians. Recently, many computer-aided diagnosis (CAD) systems have been proposed to assist this process. However, most of them do not provide the reasoning of their classification process, which may jeopardize their credibility in practical use. To overcome this, we propose a novel deep learning (DL) framework called multi-attribute attention network (MAA-Net) that is designed to mimic the clinical diagnosis process. The proposed model learns to predict nodular attributes and infer their malignancy based on these clinically-relevant features. A multi-attention scheme is adopted to generate customized attention to improve each task and malignancy diagnosis. Furthermore, MAA-Net utilizes nodule delineations as nodules spatial prior guidance for the training rather than cropping the nodules with additional models or human interventions to prevent losing the context information. Validation experiments were performed on a large and challenging dataset containing 4554 patients. Results show that the proposed method outperformed other state-of-the-art methods and provides interpretable predictions that may better suit clinical needs.

Intelligent diagnosis algorithm for thyroid nodules based on deep learning and statistical features

Thyroid nodule is one of the most common endocrine diseases in the adult population. Early screening and diagnosis of thyroid nodules is of great significance to patients’ subsequent treatment. An objective and accurate diagnosis algorithm for thyroid nodules is vital to improve the efficiency of clinical diagnosis and reduce the work pressure of doctors. In this paper, we propose a fused deep learning model for the diagnosis of benign and malignant thyroid nodules. Based on the ACR TI-RADS proposed by the American Academy of Radiology, the model first extracts 33 clinically significant statistical features of composition, echogenicity, shape, margin, and echogenic foci. After PCA dimensionality reduction, the top 4 features are fused with the feature map of EfficientNet B3. The thyroid nodules are classified into benign and malignant based on the fused feature map. This model is trained and tested on 3828 clinically collected ultrasound images of thyroid nodules. It achieves an accuracy of 96.4% and an AUC of 0.965. Compared with traditional machine learning algorithms and state-of-the-art deep learning networks, the fused model we proposed has improved classification performance and amount of parameters, and is also clinically interpretable.

Clinical study of ultrasonic evaluation of T/N staging of differentiated thyroid carcinoma using AJCC 8th staging criteria.

ObjectiveTo explore the value of ultrasound in evaluating T/N staging of differentiated thyroid carcinoma (DTC).MethodsThe clinical data of 1206 patients with DTC in our hospital from January 2018 to December 2020 were retrospectively analyzed. Cervical ultrasound was performed before surgery, and the standard ultrasound images of thyroid nodules and cervical lymph nodes I to VII were retained. According to the 8th TNM staging guidelines of AJCC DTC, the T/N stages were assessed by preoperative ultrasonic data. Then, the sensitivity, specificity, negative predicted value, positive predicted value (PPV), and diagnostic value of ultrasound T/N staging were assessed using postoperative pathological staging as the reference.ResultsUltrasonic T-stage had good consistency to pathological T stage in T4a and T4b tumors (kappa value>0.75), and moderate consistency to pathological T stage in T1, T2 and T3a tumors (kappa value between 0.4 and 0.75). ultrasonic T-stage had a sensitivity higher than 66%, except in T3b assessment (13/44, 29.5%, 95%CI: 16.1%-43.0%). All ultrasonic T-stage had specificity higher than 93%, except in T1b assessment (734/889, 82.6%, 95%CI: 80.1%-85.1%). The PPV of ultrasonic T1a to T4b was 94.3% (494/524), 61.0% (242/397), 54.4% (87/160), 34.3% (12/35), 20.3% (13/64), 100% (22/22) and 100% (4/4), respectively. The diagnostic accuracy values were 83% in T1a, 81% in T1b, 91% in T2, 98% in T3a, 93% in T3b, 99% in T4a and 100% in T4b. Nltrasonic N-stage had poor consistency to pathological N stage in any N stages (kappa value<0.3). The PPV of ultrasonic N0, N1, N1a and N1b was 61.0% (542/889), 55.2% (37/67), 48.2% (53/110), and 24.3% (34/140), respectively.ConclusionUltrasound has a good consistency and high accuracy in assessing the T-stage of DTC. However, the consistency and accuracy were poor in N-staging. It has a certain reference value in reducing excessive surgical treatment of DTC.

DMU-Net: Dual-route mirroring U-Net with mutual learning for malignant thyroid nodule segmentation

It is meaningful for radiologists to segment thyroid nodules in ultrasound images quickly and accurately using an effective segmentation algorithm. With the rise of deep learning in computer vision, many deep learning-based methods have been proposed to assist radiologists in diagnosing thyroid diseases, such as thyroid nodule classification, detection and segmentation, but there exist few methods paying attention to malignant thyroid nodule segmentation. The goal of thyroid nodule segmentation is to identify the type of thyroid nodule. However, the identification of thyroid nodule type has been relatively well developed and the identification work almost can’t bother radiologists. The more important for radiologists is to detect the inconspicuous malignant nodules precisely in ultrasonic images, avoiding radiologists confusing tissues and malignant thyroid nodules during their diagnosis. This paper proposes a deep learning-based CAD (Computer-aided diagnosis) method called Dual-route Mirroring U-Net (DMU-Net) to segment malignant thyroid nodules automatically. The method uses two subnets (U-shape subnet, inversed U-shape subnet) and three modules (pyramid attention module (PAM), margin refinement module (MRM), aggregation module (AM)) to extract contextual information of thyroid nodules and margin details in ultrasonic images. Further, the strategy of mutual learning is introduced from the natural image classification task to enhance the performance of DMU-Net. We train and evaluate our method on the self-built Malignant Thyroid Nodule Segmentation (MTNS) dataset. Finally, we compare the DMU-Net with several classical deep learning-based methods on the MTNS dataset and other public datasets. The results show our DMU-Net can achieve superior performance on these datasets.

GAN-Guided Deformable Attention Network for Identifying Thyroid Nodules in Ultrasound Images.

Early detection and identification of malignant thyroid nodules, a vital precursory to the treatment, is a difficult task even for experienced clinicians. Many Computer-Aided Diagnose (CAD) systems have been developed to assist clinicians in performing this task on ultrasonic images. Learning-based CAD systems for thyroid nodules generally accommodate both nodule detection/ segmentation and fine-grained classification for its malignancy, and prior researches often treat aforementioned tasks in separate stages, leading to additional computational costs. In this paper, we utilize an online class activation mapping (CAM) mechanism to guide the network to learn discriminative features for identifying thyroid nodules in ultrasound images, called CAM attention network. It takes nodule masks as localization cues for direct spatial attention of the classification module, thereby avoiding isolated training for classification. Meanwhile, we propose a deformable convolution module to add offsets to the regular grid sampling locations in the standard convolution, guiding the network to capture more discriminative features of nodule areas. Furthermore, we use a generative adversarial network (GAN)to ensure reliable deformations of nodules from the deformable convolution module. Our proposed CAM attention network has already achieved the 2nd place in the classification task of TN-SCUI 2020, a MICCAI 2020 Challenge with the largest set of thyroid nodule ultrasound images according to our knowledge. The further inclusion of our proposed GAN-guided deformable module allows for capturing more fine-grained features between benign and malignant nodules, and further improves the classification accuracy to a new state-of-the-art level.

A two-stage network with prior knowledge guidance for medullary thyroid carcinoma recognition in ultrasound images.

Accurate recognition of medullary thyroid carcinoma (MTC) is of great importance in medical diagnosis, as MTC is rare but second-most malignant thyroid cancers with a high case-fatality ratio.1 But there is a lower recognition rate on distinguishing MTC from other thyroid nodules in ultrasound images, even by experienced experts. This paper introduces the computer-aided method to tackle the challenge of recognizing MTC from ultrasound images, including limited MTC samples, and ambiguities among MTC, benign nodules, and papillary thyroid carcinoma (PTC). The recognition of MTC based on large MTC samples of ultrasound images has never been explored, as only one existing work presented a relevant dataset with a limited 21 MTC samples. This study proposes a novel method for primarily differentiating MTC samples from benign nodules and PTC that is the most common thyroid cancer. Our method is a two-stage schema with two important components including a cascaded coarse-to-fine segmentation network and a knowledge-based classification network. The cascaded coarse-to-fine segmentation network incorporates two U-Net++ networks for improving the segmentation results of thyroid nodules. Meanwhile, our knowledge-based classification network extracts and fuses semantic features of solid tissues and calcification for better recognizing the segmented nodules from the ultrasound images. In our experiments, dice similarity coefficient (DSC), intersection over union (IoU), precision, recall, and Hausdorff distance (HD) are adopted for evaluating the segmentation results of thyroid nodules, and accuracy, precision, recall, and F1-score are used for classificationevaluation. We present a well-annotated dataset including samples of 248 MTC, 240 benign nodules, and 239 PTC. For thyroid nodule segmentation, our designed cascaded segmentation network attains values of 0.776 DSC, 0.689 IoU, 0.778 precision, and 0.821 recall, respectively. By incorporating prior knowledge, our method achieves a mean accuracy of 82.1% in classifying thyroid nodules of MTC, PTC, and benign ones. Especially, our method gains the higher performance in recognizing MTC with an accuracy of 86.8%, compared to nearly 70% diagnosis accuracy of experienced doctors. The experimental results on our Fujian Provincial Hospital dataset further validate the efficiency of our proposedmethod. Our proposed two-stage method incorporates pipelines of thyroid nodules segmentation and classification of MTC, individually. Quantitative and qualitative results indicate that our proposed model achieves accurate segmentation of thyroid nodules. The results also validate that our learning-based framework facilitates the recognition of MTC, which gains better classification accuracy than experienced doctors.