Quantizing the Breast Imaging Reporting and Data System (BI-RADS) criteria into different categories with the single ultrasound modality has always been a challenge. To achieve this, we proposed a two-stage grading system to automatically evaluate breast tumors from ultrasound images into five categories based on convolutional neural networks (CNNs). This new developed automatic grading system was consisted of two stages, including the tumor identification and the tumor grading. The constructed network for tumor identification, denoted as ROI-CNN, can identify the region contained the tumor from the original breast ultrasound images. The following tumor categorization network, denoted as G-CNN, can generate effective features for differentiating the identified regions of interest (ROIs) into five categories: Category “3”, Category “4A”, Category “4B”, Category “4C”, and Category “5”. Particularly, to promote the predictions identified by the ROI-CNN better tailor to the tumor, refinement procedure based on Level-set was leveraged as a joint between the stage and grading stage. We tested the proposed two-stage grading system against 2238 cases with breast tumors in ultrasound images. With the accuracy as an indicator, our automatic computerized evaluation for grading breast tumors exhibited a performance comparable to that of subjective categories determined by physicians. Experimental results show that our two-stage framework can achieve the accuracy of 0.998 on Category “3”, 0.940 on Category “4A”, 0.734 on Category “4B”, 0.922 on Category “4C”, and 0.876 on Category “5”. The proposed scheme can extract effective features from the breast ultrasound images for the final classification of breast tumors by decoupling the identification features and classification features with different CNNs. Besides, the proposed scheme can extend the diagnosing of breast tumors in ultrasound images to five sub-categories according to BI-RADS rather than merely distinguishing the breast tumor malignant from benign.

The choice of an electroencephalograph (EEG) reference is a practical issue for the study of brain functional connectivity. To study how EEG reference influence functional connectivity estimation (FCE), this study compares the differences of FCE resulting from the different references such as REST (the reference electrode standardization technique), average reference (AR), linked mastoids (LM), and left mastoid references (LR). Simulations involve two parts. One is based on 300 dipolar pairs, which are located on the superficial cortex with a radial source direction. The other part is based on 20 dipolar pairs. In each pair, the dipoles have various orientation combinations. The relative error (RE) and Hamming distance (HD) between functional connectivity matrices of ideal recordings and that of recordings obtained with different references, are metrics to compare the differences of the scalp functional connectivity graph (FCG) derived from those two kinds of recordings. Lower RE and HD values imply more similarity between the two FCGs. Using the ideal recording (IR) as a standard, the results show that AR, LM and LR perform well only in specific conditions, i.e., AR performs stable when there is no upward component in sources’ orientation. LR achieves desirable results when the sources’ locations are away from left ear. LM achieves an indistinct difference with IR, i.e., when the distribution of source locations is symmetric along the line linking the two ears. However, REST not only achieves excellent performance for superficial and radial dipolar sources, but also achieves a stable and robust performance with variable source locations and orientations. Benefitting from the stable and robust performance of REST versus other reference methods, REST might best recover the real FCG of EEG. Thus, REST based FCG may be a good candidate to compare the FCG of EEG based on different references from different labs.

Introduction: Skin disease is one of the most common diseases and can affect people of all ages and races. However, the diagnosis of skin diseases via observation is a highly challenging task for both doctors and patients, and would benefit from the use of an intelligent system. Building a large benchmark with professional dermatologists is resource-intensive, and we believe that few-shot learning (FSL) methods would be helpful in solving the problem of annotated data scarcity. In this paper, we propose CDD-Net (Context Feature Fusion and Dual Attention Dermatology Net), a plug-in module for FSL clinical skin disease classification. Methods: Current FSL methods used in skin disease classification are limited to nonuniversal approaches and few disease classes. Our CDD-Net has a flexible structure, including a context feature–fusion module and dual-attention module to extract discriminating texture feature and emphasize contributive regions and channels. The context feature–fusion module localizes discriminatory texture details of skin lesions by integrating features from different layers, while the dual-attention module highlights discriminative regions via channel-wise and pixel-wise depictions based on weight vectors and restrains the contributions of irrelevant areas. We also present Derm104, a new clinical skin disease data benchmark that has significant coverage of rare diseases and reliable annotation between primary species and subspecies for better validation of our approach. Results: Our experiments validated the versatility of CDD-Net for different FSL methods and achieved an improvement in accuracy of up to 9.14 percentage points compared with the vanilla network, which can be considered state of the art. The ablation study also showed that the dual-attention module and context feature–fusion module worked efficiently in CDD-Net.

Indroduction: This study aims to develop a automated method for tooth segmentation and root canal measurement based on cone beam computed tomography (CBCT) images, providing objective, efficient, and accurate measurement results to guide and assist clinicians in root canal diagnosis grading, instrument selection, and preoperative planning.Methods: We utilizes Attention U-Net to recognize tooth descriptors, crops regions of interest (ROIs) based on the center of mass of these descriptors, and applies an integrated deep learning method for segmentation. The segmentation results are mapped back to the original coordinates and position-corrected, followed by automatic measurement and visualization of root canal lengths and angles.Results: Quantitative evaluation demonstrated a segmentation Dice coefficient of 96.33%, Jaccard coefficient of 92.94%, Hausdorff distance of 2.04 mm, and Average surface distance of 0.24 mm - all surpassing existing methods. The relative error of root canal length measurement was 3.42% (less than 5%), and the effect of auto-correction was recognized by clinicians.Discussion: The proposed segmentation method demonstrates favorable performance, with a relatively low relative error between automated and manual measurements, providing valuable reference for clinical applications.

This paper proposed a prewhitening invariance of noise space (PW-INN) as a new magnetoencephalography (MEG) source analysis method, which is particularly suitable for localizing closely spaced and highly correlated cortical sources under real MEG noise. Conventional source localization methods, such as sLORETA and beamformer, cannot distinguish closely spaced cortical sources, especially under strong intersource correlation. Our previous work proposed an invariance of noise space (INN) method to resolve closely spaced sources, but its performance is seriously degraded under correlated noise between MEG sensors. The proposed PW-INN method largely mitigates the adverse influence of correlated MEG noise by projecting MEG data to a new space defined by the orthogonal complement of dominant eigenvectors of correlated MEG noise. Simulation results showed that PW-INN is superior to INN, sLORETA, and beamformer in terms of localization accuracy for closely spaced and highly correlated sources. Lastly, source connectivity between closely spaced sources can be satisfactorily constructed from source time courses estimated by PW-INN but not from results of other conventional methods. Therefore, the proposed PW-INN method is a promising MEG source analysis to provide a high spatial-temporal characterization of cortical activity and connectivity, which is crucial for basic and clinical research of neural plasticity.