RESUMO
The automatic segmentation of intervertebral discs from medical images is an important task for an intelligent clinical system. In this study, a deep learning model based on the MultiResUNet model for the automatic segmentation of specific intervertebral discs is presented. MultiResUNet can easily segment all intervertebral discs in MRI images; however, when only certain specific intervertebral discs need to be segmented, problems with segmentation errors, misalignment, and noise occur. In order to solve these problems, a two-stage MultiResUNet model is proposed. Connected-component labeling, automatic cropping, and distance transform are used in the proposed method. The experimental results show that the segmentation errors and misalignments of specific intervertebral discs are greatly reduced, and the segmentation accuracy is increased to about 94%. The performance of the proposed method proves its usefulness for the automatic segmentation of specific intervertebral discs over other deep learning models, such as the U-Net, CNN-based, Attention U-Net, and MultiResUNet models.
RESUMO
This in vitro study evaluates the performance of a disposable vibrating-mesh nebulizer when used for 28 days. A lung model was used to simulate the breathing pattern of an adult with chronic obstructive pulmonary disease. The vibrating-mesh nebulizer was used for three treatments/day over 28 days without cleaning after each test. Results showed that the inhaled drug dose was similar during four weeks of use (p = 0.157), with 16.73 ± 4.46% at baseline and 15.29 ± 2.45%, 16.21 ± 2.21%, 17.56 ± 1.98%, and 17.13 ± 1.81%, after the first, second, third, and fourth weeks, respectively. The particle size distribution, residual drug volume, and nebulization time remained similar across four weeks of use (p = 0.110, p = 0.763, and p = 0.573, respectively). Mesh was inspected using optical microscopy and showed that approximately 50% of mesh pores were obscured after 84 runs, and light penetration through the aperture plate was significantly reduced after the 21st use (p < 0.001) with no correlation to nebulizer performance. We conclude that the vibrating-mesh nebulizer delivered doses of salbutamol solution effectively over four weeks without cleaning after each use even though the patency and clarity of the aperture plate were reduced by the first week of use.
RESUMO
There is no clear international consensus regarding the optimal medication therapy for treating Wilson's disease (WD). This study systematically reviews the effectiveness of various medication therapies in common use, specifically focusing on preliminary findings concerning the combination of a chelating agent and zinc. A systematic PubMed search was executed to locate original studies on the effectiveness of commonly used medications for WD published between January 1989 and August 2014. The results were used to conduct a systematic review of studies on combination therapies. A total of 17 combination therapy studies involving 1056 patients were reviewed. These were analyzed in terms of data on effectiveness, adverse effects, and mortality. Results from a pooled analysis indicate that combination therapies for hepatic patients were significantly less effective than the same therapies for neurological manifestations (47.1 vs. 78.6 %; pooled relative risk ratio (RR): 0.63, 95 % confidence interval CI 0.43-0.94; p = 0.02). Data from a subgroup analysis show that the combination therapy of penicillamine plus zinc sulfate resulted in a significantly higher mortality rate compared to all other combination therapy types (16.3 vs. 4.7 %; RR: 3.51, 95 % CI 1.54-8.00; p < 0.001). The use of combination therapies involving zinc and a chelator should be carefully monitored with close clinical observations and frequent biochemical tests, especially for WD patients with hepatic manifestations.
