Surface defect detection plays an essential role in industry, and it is challenging due to the following problems: 1) the similarity between defect and nondefect texture is very high, which eventually leads to recognition or classification errors and 2) the size of defects is tiny, which are much more difficult to be detected than larger ones. To address such problems, this article proposes an adaptive image segmentation network (AIS-Net) for pixelwise segmentation of surface defects. It consists of three main parts: multishuffle-block dilated convolution (MSDC), dual attention context guidance (DACG), and adaptive category prediction (ACP) modules, where MSDC is designed to merge the multiscale defect features for avoiding the loss of tiny defect feature caused by model depth, DACG is designed to capture more contextual information from the defect feature map for locating defect regions and obtaining clear segmentation boundaries, and ACP is used to make classification and regression for predicting defect categories. Experimental results show that the proposed AIS-Net is superior to the state-of-the-art approaches on four actual surface defect datasets (NEU-DET: 98.38% $\pm$ 0.03%, DAGM: 99.25% $\pm$ 0.02%, Magnetic-tile: 98.73% $\pm$ 0.13%, and MVTec: 99.72% $\pm$ 0.02%).

Accurate segmentation of 3D dental models derived from intra-oral scanners (IOS) is one of the key steps in many digital dental applications such as orthodontics and implants. However, it is difficult to accurately segment individual teeth and gums in 3D dental models due to the following problems: 1) the shape and appearance of adjacent teeth are very similar, which is easy to be misidentified; 2) the boundary between teeth and gums is often indistinct, especially in orthodontic patients with abnormalities such as missing and crowded teeth. To solve such problems, a Dual-Branch Geometric Attention Network (DBGANet) for 3D tooth segmentation is proposed, which can capture tooth geometric structure and detailed boundary information from multi-view geometric features encoded by 3D coordinates and normal vectors. The framework contains two branches, i.e., C-branch and N-branch. First, centroid-guided separable attention is designed in the C-branch to learn global context information by modeling the spatial dependencies of tooth point clouds, which can capture the overall geometric structure of teeth to better distinguish adjacent teeth with similar appearance. Then, Gaussian neighbor attention is designed in the N-branch to encode normal vectors to highlight detailed differences between geometric features at different points, which helps to refine the boundaries of teeth and gingiva for more accurate and smooth tooth segmentation. Extensive experiments on the real-patient datasets of 3D dental models demonstrate that the proposed DBGANet significantly outperforms state-of-the-art methods.

Automatic diagnosis of COVID-19 using chest CT images is of great significance for preventing its spread. However, it is difficult to precisely identify COVID-19 due to the following problems: 1) the location and size of lesions can vary greatly in CT images; 2) its unique characteristics are often imperceptible in imaging findings. To solve these problems, a Deep Dual Attention Network ( $\textrm {D}^{\textrm {2}}\textrm {ANet}$ ) is proposed for accurate diagnosis of COVID-19 by integrating dual attention modules (DAMs) with different scales of the feature extractor, where DAM can adaptively detect relevant lesion regions to extract discriminative imaging features of COVID-19. Specifically, DAM is implemented by two parallel blocks: global attention block (GAB) and local attention block (LAB), in which GAB is designed to roughly locate the infected regions from the entire image by modeling global contexts, while LAB is developed to explicitly highlight subtle differences of COVID-19 from other viral pneumonia in the infected regions by learning detailed lesion information. Experimental results on several public datasets show that $\textrm {D}^{\textrm {2}}\textrm {ANet}$ outperforms the state-of-the-art approaches in various performance metrics.

The noninvasive fetal ECG (FECG) is used to monitor fetal well-being at prenatal and intrapartum. However, it is challenging to detect the FECG signal from the abdominal electrocardiogram (ECG) due to the following problems: 1) the FECG signal is weak and often masked by noise and 2) the FECG signal is mixed or overlapped with the maternal ECG (MECG) signal. To solve such problems, a period-aware attention network (PA ² Net) is proposed for FECG detection by designing three modules, where an FECG period-aware attention module (FPAM), which is designed to suppress the noise interference by modeling the periods and features of signals, is first employed to detect FECG signals masked in noise; next, an MECG period-aware attention module, which is generated from the FPAM by the KL-divergence-based weight sharing module, can collaborate with the FPAM to search for mixed ECG signals of FECG and MECG. Finally, an antialiasing signal separation module is developed to estimate FECG signals from mixed ECG signals. Experiments conducted on the benchmarks show that the proposed PA ² Net can achieve excellent performance with the PPV of 98.83%, 99.74%, 99.67%, and 98.47% on PCDB, ADFECGDB, NIFECGDB, and LHDB databases, respectively, allowing obstetricians to use PA ² Net to monitor fetal health and diagnose diseases.

Diabetic Retinopathy (DR) is one of the leading causes of visual impairment and blindness in diabetic patients worldwide. Accurate Computer-Aided Diagnosis (CAD) systems can aid in the early diagnosis and treatment of DR patients to reduce the risk of vision loss, but it remains challenging due to the following reasons: 1) the relatively low contrast and ambiguous boundaries between pathological lesions and normal retinal regions, and 2) the considerable diversity in lesion size and appearance. In this paper, a Cross-Scale Fuzzy Holistic Attention Network (CSFHANet) is proposed for DR grading using fundus images, and it consists of two main components: Fuzzy-Enhanced Holistic Attention (FEHA) and Fuzzy Learning-based Cross-Scale Fusion (FLCSF). FEHA is developed to adaptively recalibrate the importance of feature elements by assigning fuzzy weights across both channel and spatial domains, which can enhance the model's ability to learn the features of lesion regions while reducing the interference from irrelevant information in normal retinal regions. Then, the FLCSF module is designed to eliminate the uncertainty in fused multi-scale features derived from different branches by utilizing fuzzy membership functions, producing a more comprehensive and refined feature representation from complex DR lesions. Extensive experiments on the Messidor-2 and DDR datasets demonstrate that the proposed CSFHANet exhibits superior performance compared to state-of-the-art methods.

Dental abnormality (DA) detection is of great significance to orthodontic treatment. However, it is difficult to detect abnormal teeth from the oral cavity due to the following problems: (1) The crowding dentition often overlaps with normal teeth; (2) The lesion regions are small on the tooth surface. To address such problems, a Category-Guide Attention U-Net (CGA-UNet) is proposed, where a deformable attention convolution (DAC) module is first devised to discriminate crowding teeth from normal ones by learning dentition spatial distribution information; then, a differential variable convolution (DVC) module is designed to perform pathological tooth identification by extracting the small lesion features; finally, an attentional feature fusion (AFF) module is developed to integrate the spatial information and lesion features to obtain the abnormal tooth region. Experiments conducted on the benchmark show excellent performance of CGA-UNet for dental abnormality detection, and it can further assist orthodontists in formulating orthodontic treatment plans.