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.

Objective: To compare difference in stability between two fixation modes in application of three-dimensional (3D) printed coronoid process prosthesis. Methods: 3D coronoid process prosthesis was reconstructed from computed tomography (CT) scanning acquired from volunteers with normal bony architecture of elbow. The prosthesis fixation was simulated using screws with diameter of 2.0 mm and 2.5 mm separately. Finite element analysis (FEA) was conducted at elbow flexion with different angles and the parameters including the maximum equivalent stress and the maximum displacement were evaluated. Results: Compared fixation using screw with the diameter of 2.0 mm and 2.5 mm, the maximum equivalent stress were 56.29 MPa and 39.96 MPa and the maximum displacement were 0.038 0 mm and 0.029 0 mm in 30° flexion of elbow; the maximum equivalent stress were 8.97 MPa and 8.12 MPa and the maximum displacement were 0.006 5 mm and 0.005 8 mm in 90° flexion of elbow; the maximum equivalent stress were 10.03 MPa and 9.32 MPa and the maximum displacement were 0.007 9 mm and 0.006 6 mm in 130° flexion of elbow. The maximum equivalent stress occurs at the junction of head and shaft with in both situations while the value from screw with diameter of 2.5 mm was lower. The displacement was smaller in the screw with diameter of 2.5 mm. Conclusion Compared to screws with the diameter of 2 mm, screw with the diameter of 2.5 mm shows superior stability in the fixation of 3D printed coronoid process prosthesis.

Objective To identify the GNAS gene mutation resulting in pseudohypoparathyroidism type Ⅰa (PHP-Ⅰa) in one patient. Methods The clinical data of a patient with pseudohypoparathyroidism type Ⅰa was retrospectively analyzed. All the 13 exons of GNAS were sequenced using Sanger method for the patient and the parents. The distribution of suspected causal mutation was screened in 478 healthy controls. To clarify the origin of the mutation, we performed targeted high-depth sequencing of GNAS exon harboring the mutation for the patient and the parents. Results The clinical data of the patient with the laboratory results of hypocalcaemia, hyperphosphataemia, elevated serum PTH, together with the features of AHO, conformed to the characterization of PHP-Ⅰa. The sequencing of GNAS exons identified a missense mutation (c.479G>C, p.R160P) located at exon 6 in the patient, which was absent in DNA of the parents. The mutation was not reported previously and was not found in the 478 healthy controls. We obtained about 8000-fold coverage from high-depth sequencing of DNA from peripheral blood of the patient and the parents. The disease-associated allele C identified in the patient was not observed in the parents. The number of reads with G allele (3984 reads) was roughly equal to that of C allele (4019 reads) from the targeted reanalysis of DNA of the patient. The results from high-depth sequencing indicated a de novo mutation in maternal germ cells. Conclusions We identified a new GNAS gene mutation (c.479G>C, p.R160P) caused PHP-Ia in a patient. Our results suggested the mutation was a maternal germline de novo mutation.

AIM:To explore the association between single nucleotide polymorphism in exon 33 (E33SNP) of thyroglobulin gene and Graves ’ disease ( GD) relapse after antithyroid drug ( ATD) withdrawal .METHODS:The healthy controls (232 cases) and GD patients with discontinued treatment (243 cases) were selected.According to the time of re-lapse, the GD patients were divided into A, B and C subgroups.The A group contained 77 cases of relapse within 1 year, B group contained 86 cases of relapse 1~2 years after treatment and C group contained 80 cases without recurrence within 2 years.The genotypes of E33SNP were identified by RT-PCR.The genotype ratio of thyroglobulin between control group and observation group was comparatively analyzed , and the levels of thyroid-stimulating hormone ( TSH) , free triiodothyro-nine (FT3), free thyroxine (FT4) and thyrotropin receptor antibody (TRAb), ophthalmopathy and goiter size in A , B and C subgroups in different genotype GD patients were investigated .Moreover , cumulative efficiency for patients with different genotypes in the observation group after ATD treatment within 2 years were analyzed .RESULTS:The genotype of E33SNP between observation group and control group had no significant difference , but a significant difference between A , B and C subgroups was observed (P<0.05).The levels of TSH, FT3 and FT4, and goiter size of the patients with different geno-types had no significant difference , while the TRAb levels and ophthalmopathy presented a significant difference ( P <0.05).In addition, the cumulative efficiency within 2 years for GD patients with E33SNP T/T, E33SNP T/C and E33SNP C/C genotypes was 61.8%, 42.6% and 21.3%, respectively, all with significant differences (P<0.05).CONCLU-SION:The GD patients with E33SNP C/C genotype have significantly higher TRAb level and ophthalmopathy rate than those in the patients with E33SNP C/T and E33SNP C/C genotypes, and are more likely to relapse after ATD treatment . The GD patients with E33SNP T/T genotype show a lower recurrence rate .Therefore, combination treatment or other treat-ment modalities may be more reasonable for the GD patients with E 33SNP C/C genotype.

Objective: To investigate the clinical features of autoimmune hepatitis (AIH) patients with poor response to treatment. Methods: A total of 61 AIH patients were enrolled, among whom 49 (80.33%) achieved complete response (good response group) and 12 (19.67%) had incomplete response (poor response group). The two groups were compared in terms of clinical manifestations, laboratory markers, abdominal ultrasound findings, pathological features by liver biopsy, and response to treatment. Continuous data were expressed as mean ± standard deviation (x±s), and the t-test was used for comparison between groups; categorical data were expressed as rates or percentages, and the chi-square test was used for comparison between groups; a binary logistic regression analysis was used to determine influencing factors. Results: Most patients were female in both groups, and there were no significant differences in sex ratio, mean age of onset, and general status including extrahepatic autoimmune disease between the two groups. Compared with the good response group, the poor response group had significantly higher levels of alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase (ALP), total bilirubin, immunoglobulin G, and immunoglobulin M (P < 0.05). Compared with the good response group, the poor response group had a significantly higher positive rate of autoimmune antibodies except anti-nuclear antibody (ANA), anti-smooth muscle antibody (SMA), antimitochondrial antibody (AMA), and AMA/M2 (75% vs 16.3%, P < 0.001), and there was a significant difference in the positive rate of gp210 antibody between the two groups (25% vs 0%, P < 0.01). There were significant differences between the poor response group and the good response group in the proportion of patients with liver cirrhosis (50.0 % vs 16.3%, P < 0.05) and splenomegaly (58.3% vs 22.4%, P < 0.05). The binary logistic regression analysis showed that a high serum level of ALP (odds ratio [OR] = 1.017, 95% confidence interval [CI] 1.001-1.033, P = 0.034), positive autoimmune antibodies except ANA, SMA, and AMA/M2 (OR = 70.842, 95% CI 2.132-2 354.371, P = 0.017), and liver cirrhosis (OR = 28.777, 95% CI 1.015-815.854, P = 0.049) were independent risk factors for initial treatment outcome. Conclusion A high serum level of ALP, positive autoimmune antibodies except ANA, SMA, and AMA/M2, and liver cirrhosis are closely associated with poor response in AIH patients.

A novel medical automatic image segmentation strategy based on guided filtering and multi-atlas is proposed to achieve accurate, smooth, robust, and reliable segmentation. This framework consists of 4 elements: the multi-atlas registration, which uses the atlas prior information; the label fusion, in which the similarity measure of the registration is used as the weight to fuse the warped label; the guided filtering, which uses the local information of the target image to correct the registration errors; and the threshold approaches used to obtain the segment result. The experimental results showed part among the 15 brain MRI images used to segment the hippocampus region, the proposed method achieved a median Dice coefficient of 86% on the left hippocampus and 87.4% on the right hippocampus. Compared with the traditional label fusion algorithm, the proposed algorithm outperforms the common brain image segmentation methods with a good efficiency and accuracy.
Algorithms ; Hippocampus ; anatomy & histology ; Humans ; Image Processing, Computer-Assisted ; methods ; Magnetic Resonance Imaging ; Neuroimaging ; Software

A novel medical automatic image segmentation strategy based on guided filtering and multi-atlas is proposed to achieve accurate, smooth, robust, and reliable segmentation. This framework consists of 4 elements: the multi- atlas registration, which uses the atlas prior information;the label fusion, in which the similarity measure of the registration is used as the weight to fuse the warped label;the guided filtering, which uses the local information of the target image to correct the registration errors; and the threshold approaches used to obtain the segment result. The experimental results showed part among the 15 brain MRI images used to segment the hippocampus region, the proposed method achieved a median Dice coefficient of 86%on the left hippocampus and 87.4%on the right hippocampus. Compared with the traditional label fusion algorithm, the proposed algorithm outperforms the common brain image segmentation methods with a good efficiency and accuracy.

A novel medical automatic image segmentation strategy based on guided filtering and multi-atlas is proposed to achieve accurate, smooth, robust, and reliable segmentation. This framework consists of 4 elements: the multi- atlas registration, which uses the atlas prior information;the label fusion, in which the similarity measure of the registration is used as the weight to fuse the warped label;the guided filtering, which uses the local information of the target image to correct the registration errors; and the threshold approaches used to obtain the segment result. The experimental results showed part among the 15 brain MRI images used to segment the hippocampus region, the proposed method achieved a median Dice coefficient of 86%on the left hippocampus and 87.4%on the right hippocampus. Compared with the traditional label fusion algorithm, the proposed algorithm outperforms the common brain image segmentation methods with a good efficiency and accuracy.

Objective To evaluate the prognostic factors of predicting the outcome of Graves disease (GD) after treatment with antithyroid drugs.MethodsA retrospective audit was performed for 306 consecutive patients with newly diagnosed GD.They were divided into successful and failure groups (including recurrent and non-stop subgroups )according to the treatment outcomes.Different prognostic factors after treatment with antithyroid drugs were compared and the state of thyrotropin suppression was observed as the euthyroid state at Months 3,6 and 12 respectively.ResultsAmong them,141patients (46.1％ ) were cured and 165 patients (53.9％) had treatment failures.Age at the time of diagnosis was (46 ±10) years in the successful group and (36 ±9) years in the failure group(t =3.152,P =0.002).Free T3 ( FT3 ) was (25.2 ±8.9) and ( 18.7 ±9.4) pmol/L in the failure and successful groups respectively (t =3.326,P =0.001).The FT3 to FT4 ratio and thyrotrophin receptor antibody (TRAb) levels were higher in the failure group ( t =3.331,3.389,P =0.001).Logistic regression analysis showed that thyroid size,FT3 to FT4 ratio and TRAb at the time of diagnosis were associated with failure outcomes.The ratio of continuing thyrotropin suppression in the recurrent subgroup was more than that in the successful group ( X2 =77.515,114.441,136.232,all P ＜ 0.01).ConclusionsThe GD patients with a large thyroid size and high pre-mediation levels of TRAb and FT3 to FT4 ratio are more prone to respond unfavorably to antithyroid drug treatment.And those with a large thyroid size and post-medication ophthalmopathy and continuing thyrotropin suppression have a high rate of recurrence.