Search

Objective To evaluate the lumbar multifidus muscle in patients with degenerative lumbar scoliosis (DLS) by magnetic resonance imaging (MRI), and to analyze the degeneration features of lumbar multifidus muscle in the convex and concave sides of different lumbar segments and the correlation with severity of spinal deformity. Methods A cross-sectional trial was performed on 45 DLS subjects (DLS group) and 42 healthy individual (control group) who were recruited among the public by our hospital from March 2019 to June 2021. Image J software was employed to measure and calculate the cross-sectional area (CSA), functional cross-sectional area (FCSA) and fat infiltration rate (FIR) of the multifidus muscles beside the central planes of the intervertebral discs of 5 lumbar segments, L1-L2 and L5-S1, on T2WI images. The Cobb angle of lumbar coronal scoliosis in DLS group was measured by full-length spinal X-ray series. The degeneration features of multifidus muscle in each segment of lumbar vertebrae in DLS group and their correlation with Cobb angle of lumbar scoliosis were analyzed. Results The FCSA of the convex and concave multifidus muscles at L4-L5 and L5-S1 levels were significantly smaller in the DLS group than the control group (P < 0.05). The FIR of the convex and concave multifidus muscles at L1-S1 levels were obviously higher in the DLS group than the control group (P < 0.05). The DLS group had statistically smaller FCSA in the concave side than the convex multifidus muscles at L3-L4, L4-L5 and L5-S1 levels (P < 0.05). In the DLS group, the Cobb angle was positively correlated with FCSA difference index of L4-L5 and L5-S1 levels (r=0.310, 0.559, P < 0.05). Conclusion In DLS patients, degeneration of multifidus muscle corresponding to the lower lumbar levels (L4-L5 and L5-S1) is positively correlated with the severity of scoliosis, and the condition is more serious in the concave side than the convex side. It is suggested that the degeneration of multidus muscle is not an initiating factor of DLS, but may be a compensatory factor.

Considering the special circumstances after nuclear accidents, such as continuous change of source term release rate, limited monitoring data, and variable observed noise, a source term inversion method based on the Gaussian puff dispersion model and particle filter is proposed in this paper. When there are only two sets of effective monitoring data, these data, based on 20% monitoring relative error, superimpose monitoring noise caused by particulate deposition. A particle filter was used to invert the source term release rate from the monitoring data considering stable, linearly increasing, and sinusoidally changing release rates. The error in the results relative to the true value is about 5%, and the maximum error is less than 10%.

An appropriate rescue plan can reduce the consequences and harm of a nuclear accident. Based on the combination of multi-attribute decision-making and cost-benefit analysis, alternative rescue schemes are ranked according to the rescue effect of a nuclear accident by the approximate ideal solution of multi-attribute decision-making theory. To reduce the cost of rescue and the radiation hazard of rescuers, the cost-benefit analysis method is used to select the optimal rescue plan. The model is applied to the actual nuclear accident rescue scheme selection, and the results show that the rescue scheme selected by the model based on multi-attribute decision-making not only has a better rescue effect but is also relatively low cost.

During the transportation of components related to nuclear materials, accidental chemical explosions may occur, resulting in the release of radionuclides. Effective decision-making during nuclear transport accidents, especially in cases with incomplete source information and a complex terrain, requires the rapid prediction of changes in radionuclide concentration. This paper proposes a method for predicting the concentration of radionuclides resulting from nuclear transport explosion accidents based on stacked long short-term memory (LSTM) networks. Specifically, this study considered plutonium-containing explosive transport and chemical explosion accidents under the pad surface of a hill as a research scenario. The diffusion data of radionuclide Pu-239 were simulated using the computational fluid dynamics (CFD) software OpenFOAM. Nuclide concentration and meteorological time series data of a specific area were selected for stacked LSTM network training and prediction based on geographical characteristics and population density. The proposed model, optimized using grid search, can stably achieve a mean absolute percentage error (MAPE) of less than 5% within 150 iterations for Pu-239 nuclide concentration prediction. The model is highly efficient and has significant practical value for use in nuclear emergencies.

Gamma (γ) analysis is the most important method for dose verification and dose distribution comparison. Because of the inconsistency of the coordinate system definitions and directions of different dose distributions and the inconsistent reference point settings, in γ analysis, the dose distributions should often be aligned for comparison. However, the alignment method of the current commonly used γ analysis software is extremely simple, and there are significant errors in processing the dose distribution of the large radiation field, which seriously affect the accuracy of the dose analysis and comparison. Aligning the two-dimensional dose distribution requires performing spatial transformations such as translation and rotation to achieve the same spatial position of the corresponding dose points on the two dose maps. Because the distance between different points must be constant during this alignment process, in this study, the alignment problem is transformed into a rigid registration problem. A rigid registration method based on normalized cross-correlation is used to register the dose distribution map calculated using a treatment planning system with the actually measured dose distribution map and output the corresponding translation amount after registration. 36 groups of static intensity-modulated quality assurance data designed using the pinnacle3 treatment planning system, whose planning system is consistent with the measured center point, are selected; 18 groups are small irradiation fields, and the other 18 groups are large irradiation fields. PTW VeriSoft software is used to calculate the γ passing rate under 3%/3 mm, 3%/2 mm, and 2%/2 mm pass standards separately. For small irradiation fields, the results of the direct calculation of different standard γ passing rates after alignment with VeriSoft and rigid registration are identical. Among them, under the 3%/3 mm pass standard, the results of VeriSoft alignment and rigid registration are compared with those directly calculated， the average differences in γ passing rate are 0.5% and 0.3%, respectively. For the large irradiation field, compared with the directly calculated γ passing rate results, the average differences of the γ passing rate under 3%/3 mm, 3%/2 mm, and 2%/2 mm pass standards after VeriSoft alignment are 17.1%, 23.3%, and 28.3%, respectively, with ten groups of cases all exhibiting great differences. The average differences of the γ passing rate under the 3%/3 mm, 3%/2 mm, and 2%/2 mm standards after rigid registration and direct calculation are 0.4%, 1.1%, and 2.3%, respectively, and the average difference is small. Rigid registration solves the problem of significant errors in VeriSoft’s handling of large-field dose distributions. It can be used as a method for the alignment of dose distributions in different fields when verifying the dose of radiation treatment plans.