Search

AIM:To compare the efficacy and safety of photodynamic therapy (PDT) with overlapping multiple spots and single spot for treating circumscribed choroidal hemangioma.METHODS:Twenty-two patients (22 eyes) with symptomatic circumscribed choroidal hemangioma received PDT treatment. Fourteen patients received overlapping spots (two to three spots) PDT, whereas eight patients received single-spot PDT. Laser was used at 50J/cm2 for 83s in the overlapping-spot group and 50J/cm2 for 166s in the single-spot group. Clinical examination, funduscopy, fluorescein angiography, and ultrasonography were performed at baseline and after treatment.RESULTS:The mean follow-up time was 28.5±8.0 months in the overlapping-spot group and 27.0±5.0 months in the single-spot group. Nine patients (64.2%) had their vision improved over two lines on the Snellen chart, and five patients showed stable visual acuity in the overlapping-spot group. The mean thickness of tumor decreased from 2.7±0.8mm to 1.2±0.9mm, and the mean greatest tumor linear dimension decreased from 7.4±1.5mm to 4.5±3.5mm after treatment. In the single-spot group, two patients (25%) had their vision improved over two lines on the Snellen chart, and six patients had unchanged stable vision. The mean tumor thickness in this group decreased from 2.5±0.7mm to 1.4±1.0mm, and the mean greatest tumor linear dimension decreased from 7.2±1.3mm to 4.7±3.6mm. No significant differences in visual improvement and tumor regression were found between the two groups.CONCLUSION: Overlapping-spot PDT under appropriate treatment parameters and strategies is as effective and safe as single-spot PDT for treating symptomatic circumscribed choroidal hemangioma. Improved or stabilized visual acuity was achieved as a result of tumor regression.

If the components in a component-based software system come from different sources, the characteristics of the components may be different. Therefore, evaluating the reliability of a component-based system with a fixed model for all components will not be reasonable. To solve this problem, this paper combines a single reliability growth model with an architecture-based reliability model, and proposes an optimal selecting approach. First, the most appropriate model of each component is selected according to the historical reliability data of the component, so that the evaluation deviation is the smallest. Then, system reliability is evaluated according to both the relationships among components and the using frequency of each component. As the approach takes into account the historical data and the using frequency of each component, the evaluation and prediction results are more accurate than those of using a single model.

A hybrid algorithm based on attention model (HAAM) is proposed to speed up the training of back-propagation neural networks and improve the performances. The algorithm combines the genetic algorithm with the BP algorithm based on magnified error signal. The key to this algorithm lies in the partition of the BP training process into many chips with each chip trained by the BP algorithm. The chips in the same iteration are optimized by the GA operators, and those in different iterations constitute the whole training. Therefore, the HAAM obtains the ability of searching the global optimum solution relying on these operations, and it is easy to be parallelly processed. The simulation experiments show that this algorithm can effectively avoid failure training caused by randomizing the initial weights and thresholds, and solve the slow convergence problem resulted from the Flat-Spots when the error signal becomes too small. Moreover, this algorithm improves the generalization of BP network by improving the training precision instead of adding hidden neurons.