Objective To propose a novel method based on three-dimensional depthwise separable convolution network(3D DSN)for the separation of PET images with dual tracers of 18F-FDG and 18F-FAPI.Methods A total of 120 pairs of 18F-FDG and 18F-FAPI PET images of the same patient scanned separately at different time points were collected,and the dual-tracer PET image was generated through simulation.After the image registration of PET images of two tracers for ensuring spatial position matching,the registered PET images were forward-projected to generate sinogram data,and the sinogram data of two tracers were accumulated to obtain mixed sinogram data.Subsequently,the dual-tracer PET image was reconstructed using maximum likelihood expectation maximization and input into a 3D DSN based network for image separation,thereby obtaining PET images of two single tracers.Results Compared with 3D CNN method,the proposed method increased the structure similarity index measure(SSIM)of the separated 18F-FDG images to the real 18F-FDG images by 0.87%,increased the peak signal-to-noise ratio(PSNR)by 11.8%,and reduced the normalized root mean square error(NRMSE)by 52%.The SSIM of the separated 18F-FAPI images to the real 18F-FAPI images increased by 1.1%,PSNR increased by 17.0%,and NRMSE decreased by 51%.Conclusion The proposed method can be effectively applied to simultaneous PET imaging with dual PET tracers,reducing the number of scans and costs in time and money,and providing clinical doctors more accurate and abundant diagnostic information.

Objective To propose a novel method based on three-dimensional depthwise separable convolution network(3D DSN)for the separation of PET images with dual tracers of 18F-FDG and 18F-FAPI.Methods A total of 120 pairs of 18F-FDG and 18F-FAPI PET images of the same patient scanned separately at different time points were collected,and the dual-tracer PET image was generated through simulation.After the image registration of PET images of two tracers for ensuring spatial position matching,the registered PET images were forward-projected to generate sinogram data,and the sinogram data of two tracers were accumulated to obtain mixed sinogram data.Subsequently,the dual-tracer PET image was reconstructed using maximum likelihood expectation maximization and input into a 3D DSN based network for image separation,thereby obtaining PET images of two single tracers.Results Compared with 3D CNN method,the proposed method increased the structure similarity index measure(SSIM)of the separated 18F-FDG images to the real 18F-FDG images by 0.87%,increased the peak signal-to-noise ratio(PSNR)by 11.8%,and reduced the normalized root mean square error(NRMSE)by 52%.The SSIM of the separated 18F-FAPI images to the real 18F-FAPI images increased by 1.1%,PSNR increased by 17.0%,and NRMSE decreased by 51%.Conclusion The proposed method can be effectively applied to simultaneous PET imaging with dual PET tracers,reducing the number of scans and costs in time and money,and providing clinical doctors more accurate and abundant diagnostic information.

Introductions: Facial nerve palsy (FNP) occurs in 7-10% of temporal bone fractures. The aim of this study was to review the surgical outcome of nine patients with severe to complete traumatic facial nerve (FN) injury. Methods: The patients were evaluated clinically and FNP was graded using the House Brackmann (HB) scale. High resolution computerized tomography (HRCT) of the temporal bone was used to evaluate temporal bone fractures. Transmastoid facial nerve decompression was performed and the facial nerve function was re-evaluated in subsequent follow ups. Results: There were five cases with immediate onset and four with delayed onset of FNP. Only three cases had pure temporal bone fractures, the others were associated with other life threatening injuries. The sensitivity and specificity of HRCT temporal bone to detect the obvious facial canal fracture line were 50% and 40% respectively. 75% of patients with immediate onset of HB grade VI FN palsy who were operated within a month recovered completely. Surgeries for the delayed onset FNP were performed at a mean of 70 days (range 51-94). All recovered to HB grade II-III from severe FNP. Conclusions: Our study demonstrated that transmastoid FN decompression surgery was beneficial to traumatic nerve injury. Early intervention resulted in better outcomes. However, FN function could still be salvaged even in delayed FN decompression.

Objective:To compare the consistency and detection capability of seven 2019-nCoV nucleic acid detection kits, and provide reference for detection method selection of clinical laboratory and diagnosis of new coronavirus pneumonia.Methods:Two batches of pharyngeal swab samples were collected from tenpatients with confirmed infection of 2019-nCoV and 10 suspected patients with negative 2019-nCoV test results during January 29 to February 5, 2020 in Shenzhen Luohu People′s Hospital. Seven kinds of kits were labeled as ato g and used for nucleic acid detection respectively to evaluate the consistency of the test results of the clinical samples. A 2019-nCoV positive specimen was selected and diluted to 5-concentration gradient plates (Level-1 to 5) with RNase-free water. The positive detection rate and intra-batch repeatability of different brands of kits were compared.Results:The negative and positive coincidence rates of twenty clinical samples tested by six kinds of kits were 100%, and the positive and negative coincidence rate was 8/10 and 10/10 for the other kit, respectively. The results of intra-batch repeatability showed the CVs of viral loads tested by these seven kits were all less than 5%. In the concentration range of Level-1 to 3, the detection capability for open reading frame (ORF)1ab gene of Kit b,d and f was lower than Kit a,c,e and g, and the detection capability of kit e and g was the highest (14/15). The detection capability for N gene of Kit a (15/15) was higher than the other 5 kits. The comprehensive analysis of the detection capability for ORF1ab and N gene showedthat Kit d had the lowest detection capability (ORF1ab:40%,N:53%), and there was no significant difference in the detection capability of Kit a, b, c, e, and f.Conclusions:There was no significant difference in the accuracy and repeatability of the seven kits for positive samples with high viral loads, and the detection performance was good; but some kits had poor detection capability for weak positive samples. It is suggested that the weak positive samples should be rechecked by at least two manufacturers′ kits to ensure the accuracy of the results.