Brain age prediction is of great significance to the in-depth understanding of individual neurodevelopment,early diagnosis of neuropsychiatric disorders,and formulation of personalized treatment plans. With the continuous advancement of deep learning,more and more researches focus on using such algorithms to predict brain age. Compared with traditional regression algorithms,deep learning which has the advantages of complex pattern learning,end-to-end learning and high adaptability can more accurately reveal the neuropathological mechanisms of neuropsychiatric disorders,and provide more precise tools for clinical assessment,assisted diagnosis and prognosis prediction. Herein the study reviews the recent advances in the application of deep learning algorithms in brain age prediction,introduces the achievements in deep learning model optimization,multimodal data inputs and interpretability studies for brain age prediction,discusses the methods for the establishment of integrated deep learning architectures and the future challenges of developing unified benchmarking,and provides an outlook on the application of deep learning in brain age prediction.

Brain age prediction is of great significance to the in-depth understanding of individual neurodevelopment,early diagnosis of neuropsychiatric disorders,and formulation of personalized treatment plans. With the continuous advancement of deep learning,more and more researches focus on using such algorithms to predict brain age. Compared with traditional regression algorithms,deep learning which has the advantages of complex pattern learning,end-to-end learning and high adaptability can more accurately reveal the neuropathological mechanisms of neuropsychiatric disorders,and provide more precise tools for clinical assessment,assisted diagnosis and prognosis prediction. Herein the study reviews the recent advances in the application of deep learning algorithms in brain age prediction,introduces the achievements in deep learning model optimization,multimodal data inputs and interpretability studies for brain age prediction,discusses the methods for the establishment of integrated deep learning architectures and the future challenges of developing unified benchmarking,and provides an outlook on the application of deep learning in brain age prediction.

Aim To investigate whether nicorandil (Nic)protects H9c2 cardiac cells against high glucose (HG)-induced injury and inflammation by inhibiting nuclear factor-κB (NF-κB )/cyclooxygenase-2 (COX-2 )pathway.Methods Cell viability was measured by cell counter kit-8 (CCK-8)assay.The expression lev-els of NF-κB,COX-2 and cleaved caspase-3 were de-termined by Western blot.The activity of lactate dehy-drogenase (LDH)in the culture medium was measured with commercial kits.The intracellular level of reactive oxygen species (ROS)was detected by 2′,7′-dichlor-fluorescein-diacetate (DCFH-DA)staining followed by photofluorography.The number of apoptotic cells was observed by Hoechst 33258 nuclear staining followed by photofluorography.Mitochondrial membrane poten-tial (MMP)was examined by rhodamine 123 staining followed by photofluorography.The secretion levels of interleukin-1β(IL-1β) and tumor necrosis factor-α(TNF-α) were detected by ELISA.Results After H9 c2 cardiac cells were treated with 35 mmol · L-1 glucose (high glucose,HG)for 24 h,the cell viability was significantly decreased .Pre-treatment of the cells with 20~100 μmol·L-1 Nic for 60 min or 50 μmol· L-1 Nic for 30~120 min before exposure to HG signif-icantly attenuated the decrease in viability induced by HG.On the other hand,HG increased the expression levels of phosphorated (p)-NF-κB p65 and cyclooxy-genase-2 (COX-2 )in H9c2 cardiac cells.Pre-treat-ment of the cells with 50 μmol·L-1 Nic for 60 min at-tenuated the up-regulation of p-NF-κB p65 and COX-2 expression levels induced by HG.Furthermore,HG induced considerable injuries and inflammatory re-sponse,leading to increases in LDH activity,ROS generation,MMP loss,the number of apoptotic cells, the expression of cleaved caspase-3 as well as the se-cretion levels of IL-1βand TNF-α.Pre-treatment of the cells with 50 μmol·L-1 Nic for 60 min before HG exposure,or co-treatment of the cells with 100 μmol· L-1 PDTC (an inhibitor of NF-κB)or 10 μmol·L-1 NS-398 (an inhibitor of COX-2)and HG for 24 h ob-viously reduced the above injuries and inflammatory re-sponse induced by HG. Conclusion Nic protects H9 c2 cardiac cells against HG-induced injury and in-flammation by inhibiting NF-κB/COX-2 pathway.

Objective To investigate the prevalence and risk factors of chronic kidney disease (CKD) in the adult urban population of Hezhou Guangxi. Methods One thousand and two hundred urban residents (older than 18 years) from Hezhou Guangxi were randomly selected using a random sampling. All the residents were interviewed. Their morning spot urine were tested to determine albumin to ereatinine ratio (abnormal:≥30 mg/g), and renal function [abnomal: eMDRD <60 ml·min-1·(1.73 m2)-1] was assessed. Morning spot urine dipstick of hematuria (abnormal:≥1 +) was confirmed by microscopy (abnormal: 3 red blood cells/HP). The associations among demographic characteristics, health eharacteristies and indicators of kidney damage were examined. Results Eligible data of 1069 subjects were enrolled in the study. The prevalence of albuminuria was 7.5%, hematuria 4.8%, and reduced eGFR 3.6%. The prevalence of kidney disease was 14.4% and the recognition was 1.4%. Age (OR 1.022, 95%CI 1.008-1.035), gender (OR 2.249, 95%CI 1.502-3.367), diabetes mellitus (OR 7.422, 95%CI 3.985-13.825) and hypertension (OR 4.397, 95% CI 2.601-7.432) were independently associated with CKD. Conclusions The prevalence of chronic kidney disease is 14.4% and the recognition is 1.4% in adult urban population of Hezhou Guangxi. Independent risk factors associated with chronic kidney disease are age, gender, diabetes mellitus and hypertension which is similar to those in developed countries and domestic big cities.

AimTo study the effects of glipizide and met formin on the serum IGF-1,IGF-2 in patients with type Ⅱ diabetes mellitus; Methods The effect of glipizide(n ＝ 40) and metformin(n ＝ 25) on serum IGF-1, IGF-2 in patients with type Ⅱ diabetes mellitus were compared with self- controlled study. Results In metformin-treated patients ,there were not significantly changes in fasting IGF-1 and IGF-2 concentrations, In glipizide-treated patients, there were markedly increased IGF-1 concentrations(181.8+ 104.5) vs (209.0+ 88.2) ng· ml-1(P＜0.05) while serum IGF-2 was not change. There was a significant reduction of blood glucose in two groups at the end of treatment(both P＜0.01), but C-peptide level was markedly increased(P＜0.05) only in glipizide-treatedpatients.Conclusion The changes of IGF-1 is markedly different between metformin-treated and glipizide-treated patients with type Ⅱ diabetes mellitus.