The deep convolutional neural network (ConvNet) achieves significant segmentation performance on medical images of various modalities. However, the isolated errors in a large testing set with various tumor conditions are not acceptable in clinical practice. This is usually caused in inadequate training and noise inherent during data collection, which are recognized as epistemic and aleatoric uncertainties in deep learning-based approaches. In this paper, we analyze the two types of uncertainties in medical image segmentation tasks and propose a reduction method by training models with data augmentation. The shelter zones in images are reduced with 2D imaging on surfaces of different angles from 3D organs. Rotation transformation and noise are estimated by Monte Carlo simulation with prior parameter distributions, and the aleatoric uncertainty is quantized in this process. Experiments on segmentation of computed tomography images demonstrate that overconfident incorrect predictions are reduced through uncertainty reduction and that our method outperforms prediction baselines based on epistemic and aleatoric estimation.
The software-defined (SD) and edge computing (EC) are emerging technologies that have been used to improve the network operation efficiency of wireless sensor networks (WSNs). Due to the advantages of the SD and EC technologies, the area of WSNs has achieved a new dimension and breakthrough. However, the limited energy allocation mechanism in edge-SD wireless sensor networks (ESDWSNs) makes the energy consumption of different nodes unbalanced. In this paper, we propose an energy allocation optimization (EAO) algorithm that solves the energy averaging and minimization (ECAM) problem in ESDWSNs by selecting appropriate relay nodes and de-duplicated data flows. Specifically, we first establish a novel three-layer network architecture based on the edge computing and software-defined technologies. Then we proposed the ECAM problem, which minimizes the energy consumption in ESDWSNs. Furthermore, we propose an adaptive Levenberg-Marquardt algorithm and derive the optimization value of energy cost function. The extensive simulation results based on the NS-2 simulator demonstrate the energy balance efficiency of the EAO algorithm in ESDWSNs.