Assessing the location and extent of lesions caused by chronic stroke is critical for medical diagnosis, surgical planning, and prognosis. In recent years, with the rapid development of 2D and 3D convolutional neural networks (CNN), the encoder-decoder structure has shown great potential in the field of medical image segmentation. However, the 2D CNN ignores the 3D information of medical images, while the 3D CNN suffers from high computational resource demands. This paper proposes a new architecture called dimension-fusion-UNet (D-UNet), which combines 2D and 3D convolution innovatively in the encoding stage. The proposed architecture achieves a better segmentation performance than 2D networks, while requiring significantly less computation time in comparison to 3D networks. Furthermore, to alleviate the data imbalance issue between positive and negative samples for the network training, we propose a new loss function called Enhance Mixing Loss (EML). This function adds a weighted focal coefficient and combines two traditional loss functions. The proposed method has been tested on the ATLAS dataset and compared to three state-of-the-art methods. The results demonstrate that the proposed method achieves the best quality performance in terms of DSC = 0.5349±0.2763 and precision = 0.6331±0.295).

For Si anode materials used for lithium ion batteries (LIBs), developing an effective solution to overcome their drawbacks of large volume change and poor electronic conductivity is highly desirable. Here, the composites of ZnO-incorporated and carbon-coated silicon/porous-carbon nanofibers (ZnO-Si@C-PCNFs) are designed and synthesized via a traditional electrospinning method. The prepared ZnO-Si@C-PCNFs can obviously overcome these two drawbacks and provide excellent LIB performance with excellent rate capability and stable long cycling life of 1000 cycles with reversible capacity of 1050 mA h g −1 at 800 mA g −1 . Meanwhile, anodes of ZnO-Si@C-PCNFs attached with Ag particles display enhanced LIB performance, maintaining an average capacity of 920 mA h g −1 at a large current of 1800 mA g −1 even for 1000 cycles with negligible capacity loss and excellent reversibility. In addition, the assembling method with important practical significance for a simple pouch full cell is designed and used to evaluate the active materials. The Ag/ZnO-Si@C-PCNFs are prelithiated and assembled in full cells using LiNi 0.5 Co 0.2 Mn 0.3 O 2 (NCM523) as cathodes, exhibiting higher energy density (230 W h kg −1 ) of 18% than that of 195 W h kg −1 for commercial graphite//NCM523 full pouch cells. Importantly, the comprehensive mechanisms of enhanced electrochemical kinetics originating from ZnO-incorporation and Agattachment are revealed in detail.

Segmenting stroke lesions from T1-weighted MR images is of great value for large-scale stroke rehabilitation neuroimaging analyses. Nevertheless, there are great challenges with this task, such as large range of stroke lesion scales and the tissue intensity similarity. The famous encoder-decoder convolutional neural network, which although has made great achievements in medical image segmentation areas, may fail to address these challenges due to the insufficient uses of multi-scale features and context information. To address these challenges, this paper proposes a Cross-Level fusion and Context Inference Network (CLCI-Net) for the chronic stroke lesion segmentation from T1-weighted MR images. Specifically, a Cross-Level feature Fusion (CLF) strategy was developed to make full use of different scale features across different levels; Extending Atrous Spatial Pyramid Pooling (ASPP) with CLF, we have enriched multi-scale features to handle the different lesion sizes; In addition, convolutional long short-term memory (ConvLSTM) is employed to infer context information and thus capture fine structures to address the intensity similarity issue. The proposed approach was evaluated on an open-source dataset, the Anatomical Tracings of Lesions After Stroke (ATLAS) with the results showing that our network outperforms five state-of-the-art methods. We make our code and models available at https://github.com/YH0517/CLCI_Net.

No abstract