Diagnosis of Benign and Malignant Thyroid Nodules Using Combined Conventional Ultrasound and Ultrasound Elasticity Imaging

On the identification of thyroid nodules using semi‐supervised deep learning

“…The study in Reference 6 developed a multitask cascade convolution neural network (MC‐CNN) framework to exploit the context information of thyroid nodules. In Reference 7, the authors present a method for diagnosing benign and malignant thyroid nodules using combined conventional US and US elasticity imaging based on a convolutional neural network. The article in Reference 8 aims to compare the performance of radiomics and deep learning based methods for the classification of thyroid nodules from ultrasound images.…”

RadCloud—An Artificial Intelligence-Based Research Platform Integrating Machine Learning-Based Radiomics, Deep Learning, and Data Management

“…DL-Based Medical Image Analysis Module in the RadCloud platform has started being used for clinical researches. For example, transfer learning by tunning pretrained deep convolution neural network (CNN) models (VGG16, ResNet 18, GoogleNet, inception-V3 and AlexNet) has been used for the classification of benign and malignant thyroid nodules based on hybrid features of conventional US and US elasticity imaging[32]. 1156 thyroid nodule US images in 233 patients were included in this study, resulting in 539 benign images and 617 malignant images in total[32].…”

“…For example, transfer learning by tunning pretrained deep convolution neural network (CNN) models (VGG16, ResNet 18, GoogleNet, inception-V3 and AlexNet) has been used for the classification of benign and malignant thyroid nodules based on hybrid features of conventional US and US elasticity imaging[32]. 1156 thyroid nodule US images in 233 patients were included in this study, resulting in 539 benign images and 617 malignant images in total[32]. U-net based deep CNN models integrated in DL module have been used for automatic segmentation of Type B aortic dissection (TBAD) from computed tomography angiography (CTA)…”

Statistical analysis and machine learning in multimodal brain imaging of neuropsychiatric disorders

“…The deep model architecture and nonlinear layers of CNN also allow investigating the complicated data patterns of images (Lee et al, 2017). Nevertheless, the application of deep learning to classify Qin et al, 2020a), causing a lack of applications for psychiatric disorder classification. Recent works on whole-brain MRI images have utilized 3D CNN models for classifying schizophrenia patients and healthy controls (Hu et al, 2020) (Oh et al, 2020).…”

“…Deep feature method was utilized for relatively small datasets and deep features were extracted from the frozen pre-trained networks then classified with fully-connected layers or conventional classifiers such as SVM and logistic regression (LR) (Cheplygina et al, 2019;Morid et al, 2020). A few studies performed extra feature selection via pooling layers (Song et al, 2019;Zhu et al, 2019), or statistical methods (Qin et al, 2020b) to reduce the high dimensionality of extracted features. Most studies reported superior or comparable performance comparing to handcrafted feature-based machine learning.…”