Automated Segmentation Method for Low Field 3D Stomach MRI Using Transferred Learning Image Enhancement Network

Accurate segmentation of abdominal organs has always been a difficult problem, especially for organs with cavities. And MRI-guided radiotherapy is particularly attractive for abdominal targets compared with low CT contrast. But in the limit of radiotherapy environment, only low field MRI segmentation can be used for stomach location, tracking, and treatment planning. In clinical applications, the existing 3D segmentation network model is trained by the low field MRI, and the segmentation result cannot be used in radiotherapy plan since the bad segmentation performance. Another way is that historical high field intensity MR images are directly used for data expansion to network learning; there will be a domain shift problem. How to use different domain images to improve the segmentation accuracy of deep neural network? A 3D low field MRI stomach segmentation method based on transfer learning image enhancement is proposed in this paper. In this method, Cycle Generative Adversarial Network (CycleGAN) is used to construct and learn the mapping relationship between high and low field intensity MRI and to overcome domain shift. Then, the image generated by the high field intensity MRI through the CycleGAN network is with transferred information as the extended data. The low field MRI combines these extended datasets to form the training data for training the 3D Res-Unet segmentation network. Furthermore, the convolution layer, batch normalization layer, and Relu layer together were replaced with a residual module to relieve the gradient disappearance of the neural network. The experimental results show that the Dice coefficient is 2.5 percent better than the baseline method. The over segmentation and under segmentation are reduced by 0.7 and 5.5 percent, respectively. And the sensitivity is improved by 6.4 percent.

Numerical Modeling of Photonic Jet behind Triangular Prism

The scattering of electromagnetic plane waves by triangular prism and its truncated form (the isosceles triangle and the trapezoid are transverse sections, respectively) has been studied in order to determine possibility of high field intensity (photonic jet) formation. Using high-resolution finite-difference time-domain simulation, an optimal relationship between the wavelength and the size of the prism was found to form photonic jet with sub wavelength waist on the shadow side of the prism. Truncation of the prism (with trapezoids as transverse sections) leads to an improvement in the characteristics of photonic jets (intensity, length and waist). A qualitative explanation of the simulation results obtained is presented.

Optimal Canceling of the Physiological Tremor for Rehabilitation in Parkinson’s Disease

Introduction: This study was conducted to control hand tremors and decrease adverse effects due to the high field intensity in advanced Parkinson’s disease. We aimed at concurrently controlling two areas of Basal Ganglia (BG) in a closed-loop strategy. Methods: In the present research, two nuclei of BG, namely subthalamic nucleus and globus pallidus internal were simultaneously controlled. Furthermore, to enhance the feasibility of the suggested control strategy, the coefficients of the controller were determined using a hybrid version of the harmony search and cuckoo optimization algorithm. Results: The advantages of the applied method include decreasing hand tremors and applied electric field intensity to the brain; consequently, it leads to reducing adverse effects, such as muscle contraction and speech disorders. Moreover, the purposed controller has achieved superior performance against changing the parameters of the model (robustness analysis) and under noise tests, compared to other conventional controllers, such as Proportional Integrator (PI) and Proportional Derivative (PD). Conclusion: The employed approach provided an effective strategy to reduce hand tremors. It also decreased the delivered high field intensity to the brain; consequently, it reduced adverse effects, such as memory loss and speech disorders. It is important to ascertain the superior performance of the suggested closed-loop control scheme in different conditions and levels of tremor. Such a function was examined in terms of robustness against the variation of parameters and uncertainties. We also obtained time domain outcomes, i.e., compared with the state-of-the-art approaches.