Focal white matter microstructural alteration after anthracycline-based systemic treatment in long-term breast cancer survivors: a structural magnetic resonance imaging study.

Understanding the neural correlates of cognitive problems in patients with breast cancer (BC) after systemic treatment have been a topic of increasing investigation. The heterogeneity of the systemic treatment regimens may undermine our ability to identify brain microstructural alterations resulting from any given regimen. We investigated the detrimental effects of the anthracycline-based systemic treatment (AST) regimen (epirubicin and cyclophosphamide + docetaxel + tamoxifen) on brain gray matter (GM) and white matter (WM) microstructural alteration in long-term BC survivors. We performed a battery of neuropsychological tests and structural magnetic resonance imaging (MRI) to 31 long-term BC survivors who had received the AST regimen (AST group) and 43 healthy controls (HC group). Voxel-based morphometry evaluated the whole-brain voxel-wise GM volume, while diffusion tensor imaging technique with tract-based spatial statistics analysis evaluated whole-brain WM microstructural alteration. Partial least squares regression (PLSR) was used to evaluate the relationship between cognitive impairment and brain microstructural alteration in BC survivors. Compared with the HC group, the AST group exhibited a significantly poorer performance in attention, as well as a marginal significantly poorer performance in verbal working memory and executive function. Significantly lower fractional anisotropy (FA), higher radial diffusivity (RD), and lower axial diffusivity (AD) in multiple brain WM regions were showed in AST group compared with the HC group. Overlap of lower FA and higher RD was found in the body of corpus callosum (CC) and bilateral superior corona radiata (SCR), whereas overlap of lower FA and AD was found in the body of CC and right SCR. The PLSR results showed that the WM regions with overlap of lower FA and AD were significantly associated with executive and verbal working memory decline. No significant difference was observed in the GM volume between groups. Our results suggest that microstructural abnormalities of certain vulnerable WM regions in the AST regimen-exposed brain may provide neuroimaging evidence for the identification of brain injury and cognitive impairment induced by specific chemotherapy regimens.

High Quality Coal Foreign Object Image Generation Method Based on StyleGAN-DSAD.

Research on coal foreign object detection based on deep learning is of great significance to safe, efficient, and green production of coal mines. However, the foreign object image dataset is scarce due to collection conditions, which brings an enormous challenge to coal foreign object detection. To achieve augmentation of foreign object datasets, a high-quality coal foreign object image generation method based on improved StyleGAN is proposed. Firstly, the dual self-attention module is introduced into the generator to strengthen the long-distance dependence of features between spatial and channel, refine the details of the generated images, accurately distinguish the front background information, and improve the quality of the generated images. Secondly, the depthwise separable convolution is introduced into the discriminator to solve the problem of low efficiency caused by the large number of parameters of multi-stage convolutional networks, to realize the lightweight model, and to accelerate the training speed. Experimental results show that the improved model has significant advantages over several classical GANS and original StyleGAN in terms of quality and diversity of the generated images, with an average improvement of 2.52 in IS and a decrease of 5.80 in FID for each category. As for the model complexity, the parameters and training time of the improved model are reduced to 44.6% and 58.8% of the original model without affecting the generated images quality. Finally, the results of applying different data augmentation methods to the foreign object detection task show that our image generation method is more effective than the traditional methods, and that, under the optimal conditions, it improves APbox by 5.8% and APmask by 4.5%.