We proposed a new stitching method based on sift features to obtain an enlarged view of transmission electron microscopic (TEM) images with a high resolution. The sift features were extracted from the images, which were then combined with fitted polynomial correction field to correct the images, followed by image alignment based on the sift features. The image seams at the junction were finally removed by Poisson image editing to achieve seamless stitching, which was validated on 60 local glomerular TEM images with an image alignment error of 62.5 to 187.5 nm. Compared with 3 other stitching methods, the proposed method could effectively reduce image deformation and avoid artifacts to facilitate renal biopsy pathological diagnosis.

We proposed a new stitching method based on sift features to obtain an enlarged view of transmission electron microscopic (TEM) images with a high resolution. The sift features were extracted from the images, which were then combined with fitted polynomial correction field to correct the images, followed by image alignment based on the sift features. The image seams at the junction were finally removed by Poisson image editing to achieve seamless stitching, which was validated on 60 local glomerular TEM images with an image alignment error of 62.5 to 187.5 nm. Compared with 3 other stitching methods, the proposed method could effectively reduce image deformation and avoid artifacts to facilitate renal biopsy pathological diagnosis.

We proposed a new stitching method based on sift features to obtain an enlarged view of transmission electron microscopic (TEM) images with a high resolution. The sift features were extracted from the images, which were then combined with fitted polynomial correction field to correct the images, followed by image alignment based on the sift features. The image seams at the junction were finally removed by Poisson image editing to achieve seamless stitching, which was validated on 60 local glomerular TEM images with an image alignment error of 62.5 to 187.5 nm. Compared with 3 other stitching methods, the proposed method could effectively reduce image deformation and avoid artifacts to facilitate renal biopsy pathological diagnosis.
Algorithms ; Artifacts ; Humans ; Image Processing, Computer-Assisted ; methods ; Kidney Glomerulus ; ultrastructure ; Microscopy, Electron, Transmission ; methods

Objective:To explore the risk factors of ischemic stroke after coronary artery bypass grafting(CABG) in elderly(≥75 years old)patients, establish a risk prediction model and evaluate it.Methods:From January 2015 to September 2021, a total of 1 553 elderly patients with coronary artery disease who were admitted to Beijing Anzhen Hospital for coronary artery bypass grafting were included retrospectively. Among which 1 121(72%) cases were males, with a median age of 77( IQR 75, 78) years. Clinical data were collected and univariate analysis and multiple logistic regression analysis were used to explore the risk factors of ischemic stroke after CABG in elderly patients. After the establishment of risk prediction model, we constructed the nomogram, and tested the discrimination and calibration of the model. Results:All patients underwent CABG, there were 35 patients with ischemic stroke after operation, with an incidence of 2.25%(35/1 553). Multivariate logistic regression analysis showed that diabetes( OR=2.61, 95% CI: 1.31-5.32), old myocardial infarction( OR=3.62, 95% CI: 1.61-7.63), systolic blood pressure( OR=1.03, 95% CI: 1.01-1.04) and vertebral artery stenosis( OR=1.01, 95% CI: 1.00-1.02) were independent risk factors for postoperative cerebral infarction in patients undergoing CABG. The model was presented by a nomogram, and the model discrimination was evaluated by ROC curve. The area under the curve( AUC) was 0.757, indicating a optimal discrimination. Hosmer- Lemeshow test of goodness of fit was performed to evaluate the model calibration( χ2=6.209, P=0.624). Conclusion:Diabetes mellitus, old myocardial infarction, systolic blood pressure and vertebral artery stenosis are independent risk factors for ischemic stroke in elderly patients after CABG. The established risk prediction model has optimal discrimination and calibration.

Lung cancer is one of the most harmful global diseases with high morbidity and mortality rate. As a unique kind of driver gene, the pathogenic fusion gene is a common mechanism of lung cancer. Most fusion genes are produced by chromosome rearrangement and encoding receptor tyrosine kinases, which could be potential lung cancer therapeutic targets. Since ALK was first identified in 2007, methods like FISH, IHC, RT-PCR and NGS have been intensively applied, leading to identification of multiple other lung cancer fusion genes including ROS1, RET, FGFR, NTRK1, NRG1, DNAH5 and LTK. These works broaden the spectrum of lung cancer related gene mutations, and support the customized treatment for clinical patients. For some fusion genes, corresponding kinase inhibitors have been developed with good efficacy, however, the treatment is still being challenged by several problems like drug resistance. Based on recent studies, the research development of lung cancer fusion genes will be discussed.

OBJECTIVE To establish a method for determining the content of 11 antioxidants and their degradation products in recombinant Exendin-4-FC fusion protein injection by HPLC. METHODS The protein was precipitated with saturated ammonium sulfate. After centrifugation, the supernatant was transferred to a C18 solid phase extraction cartridge activated by methanol. Then the cartridge was eluted with 4 mL of methanol and 5 mL of ethyl acetate respectively, and the eluent was diluted with methanol-ethyl acetate(2∶3) mixed solvent and passed through a 0.22 µm PTFE hydrophobic filter. It was analyzed by HPLC and quantified by external standard method. Chromatographic conditions: Kinetex® XB-C18 100Å (100 mm×4.6 mm, 2.6 µm)column, the detection wavelength was 230 nm, the column oven was 30 ℃, the injection volume was 5 µL and the flow rate was 0.4 mL·min–1, mobile phase was 0.1% formic acid-methanol(A)-0.1% formic acid aqueous solution(B), the running time was 45 min. RESULTS The 11 target substances showed a good linear relationship in the range of 2.5−35 μg·mL–1 with R2 ≥0.99. At three different concentration(25, 10, 5 μg·mL–1) of spiked samples, the average recovery rates of 11 antioxidants ranged from 88.1% to 106.5%, with RSDs in the range of 0.10%–9.05%. The RSDs of 6 repeatable samples was 2.01%–4.77%, which of 12 intermediate precision samples was 2.58%–9.75%. The positive/inverted samples of three batches of recombinant Exendin-4-FC fusion protein injection were detected at 0 month, 3 months and 6 months(25 ℃), and the results showed that there was no antioxidant and its degradation leaching in all batches of samples at different detection points. CONCLUSION The method has good specificity, high accuracy and precision, good solution stability, high durability and can be used for the content detection of antioxidants in drugs.