Payment Channel Networks (PCNs) are a promising alternative to improve the scalability of a blockchain network. A PCN employs off-chain micropayment channels that do not need a global block confirmation procedure, thereby sacrificing the ability to confirm transactions instantaneously. PCN uses a routing algorithm to identify a path between two users who do not have a direct channel between them to settle a transaction. The performance of most of the existing centralized path-finding algorithms does not scale with network size. The rapid growth of Bitcoin PCN necessitates considering distributed algorithms. However, the existing decentralized algorithms suffer from resource underutilization. We present a decentralized routing algorithm, Swift, focusing on fee optimization. The concept of a secret path is used to reduce the path length between a sender and a receiver to optimize the fees. Furthermore, we reduce a network structure into combinations of cycles to theoretically study fee optimization with changes in cloud size. The secret path also helps in edge load sharing, which results in an improvement of throughput. Swift routing achieves up to 21% and 63% in fee and throughput optimization, respectively. The results from the simulations follow the trends identified in the theoretical analysis.
Magnetic resonance imaging (MRI) has played an important role in the rapid growth of medical imaging diagnostic technology, especially in the diagnosis and treatment of brain tumors owing to its non-invasive characteristics and superior soft tissue contrast. However, brain tumors are characterized by high non-uniformity and non-obvious boundaries in MRI images because of their invasive and highly heterogeneous nature. In addition, the labeling of tumor areas is time-consuming and laborious.
To address these issues, this study uses a residual grouped convolution module, convolutional block attention module, and bilinear interpolation upsampling method to improve the classical segmentation network U-net. The influence of network normalization, loss function, and network depth on segmentation performance is further considered.
In the experiments, the Dice score of the proposed segmentation model reached 97.581%, which is 12.438% higher than that of traditional U-net, demonstrating the effective segmentation of MRI brain tumor images.
In conclusion, we use the improved U-net network to achieve a good segmentation effect of brain tumor MRI images.