Although breast ultrasonography is the mainstay modality for differentiating between benign and malignant breast masses, it has intrinsic problems with false positives and substantial interobserver variability. Artificial intelligence (AI), particularly with deep learning models, is expected to improve workflow efficiency and serve as a second opinion. AI is highly useful for performing three main clinical tasks in breast ultrasonography: detection (localization/ segmentation), differential diagnosis (classification), and prognostication (prediction). This article provides a current overview of AI applications in breast ultrasonography, with a discussion of methodological considerations in the development of AI models and an up-to-date literature review of potential clinical applications.
The use of three-dimensional face-scanning systems to obtain facial models is of increasing interest, however, systematic assessments of the reliability of portable face-scan devices have not been widely conducted. Therefore, a systematic review and meta-analysis were performed considering the accuracy of facial models obtained by portable face-scanners in comparison with that of those obtained by stationary face-scanning systems. A systematic literature search was conducted in electronic databases following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines for articles published from 1 January 2009 to 18 March 2020. A total of 2806 articles were identified, with 21 articles available for the narrative review and nine studies available for meta-analysis. The meta-analysis revealed that the accuracy of the digital face models generated by the portable scanners was not significantly different from that of the stationary face-scanning systems (standard mean difference (95% confidence interval) = −0.325 mm (−1.186 to 0.536); z = −0.74; p = 0.459). Within the comparison of the portable systems, no statistically significant difference was found concerning the accuracy of the facial models among scanning methods (p = 0.063). Overall, portable face-scan devices can be considered reliable for obtaining facial models. However, caution is needed when applying face-scanners with respect to scanning device settings, control of involuntary facial movements, landmark and facial region identifications, and scanning protocols.
To learn spatiotemporal representations and anomaly predictions from geophysical data, we propose STANet, a spatiotemporal neural network with a trainable attention mechanism, and apply it to El Niño predictions for long-lead forecasts. The STANet makes two critical architectural improvements: it learns spatial features globally by expanding the network’s receptive field and encodes long-term sequential features with visual attention using a stateful long-short term memory network. The STANet conducts multitask learning of Nino3.4 index prediction and calendar month classification for predicted indices. In a comparison of the proposed STANet performance with the state-of-the-art model, the accuracy of the 12-month forecast lead correlation coefficient was improved by 5.8% and 13% for Nino3.4 index prediction and corresponding temporal classification, respectively. Furthermore, the spatially attentive regions for the strong El Niño events displayed spatial relationships consistent with the revealed precursor for El Niño occurrence, indicating that the proposed STANet provides good understanding of the spatiotemporal behavior of global sea surface temperature and oceanic heat content for El Niño evolution.
15 April 2019 in IEEE Transactions on Medical Imaging
Missing data is a common problem in longitudinal studies due to subject dropouts and failed scans. We present a graph-based convolutional neural network to predict missing diffusion MRI data. In particular, we consider the relationships between sampling points in the spatial domain and the diffusion wave-vector domain to construct a graph. We then use a graph convolutional network to learn the non-linear mapping from available data to missing data. Our method harnesses a multi-scale residual architecture with adversarial learning for prediction with greater accuracy and perceptual quality. Experimental results show that our method is accurate and robust in the longitudinal prediction of infant brain diffusion MRI data.
The early and accurate diagnosis of skin cancer is crucial for providing patients with advanced treatment by focusing medical personnel on specific parts of the skin. Networks based on encoder–decoder architectures have been effectively implemented for numerous computer-vision applications. U-Net, one of CNN architectures based on the encoder–decoder network, has achieved successful performance for skin-lesion segmentation. However, this network has several drawbacks caused by its upsampling method and activation function. In this paper, a fully convolutional network and its architecture are proposed with a modified U-Net, in which a bilinear interpolation method is used for upsampling with a block of convolution layers followed by parametric rectified linear-unit non-linearity. To avoid overfitting, a dropout is applied after each convolution block. The results demonstrate that our recommended technique achieves state-of-the-art performance for skin-lesion segmentation with 94% pixel accuracy and a 88% dice coefficient, respectively.