RESUMO
Resting-state functional magnetic resonance imaging (fMRI) provides an efficient way to analyze the functional connectivity between brain regions. A comprehensive understanding of brain functionality requires a unified description of multi-scale layers of neural structure. However, existing brain network modeling methods often simplify this property by averaging Blood oxygen level dependent (BOLD) signals at the brain region level for fMRI-based analysis with the assumption that BOLD signals are homogeneous within each brain region, which ignores the heterogeneity of voxels within each Region of Interest (ROI). This study introduces a novel multi-stage self-supervised learning framework for multiscale brain network analysis, which effectively delineates brain functionality from voxel to ROIs and up to sample level. A Contrastive Voxel Clustering (CVC) module is proposed to simultaneously learn the voxel-level features and clustering assignments, which ensures the retention of informative clustering features at the finest voxel-level and concurrently preserves functional connectivity characteristics. Additionally, based on the extracted features and clustering assignments at the voxel level by CVC, a Brain ROI-based Graph Neural Network (BR-GNN) is built to extract functional connectivity features at the brain ROI-level and used for sample-level prediction, which integrates the functional clustering maps with the pre-established structural ROI maps and creates a more comprehensive and effective analytical tool. Experiments are performed on two datasets, which illustrate the effectiveness and generalization ability of the proposed method by analyzing voxel-level clustering results and brain ROIs-level functional characteristics. The proposed method provides a multiscale modeling framework for brain functional connectivity analysis, which will be further used for other brain disease identification. Code is available at https://github.com/yanliugroup/fmri-cvc.
Assuntos
Encéfalo , Imageamento por Ressonância Magnética , Rede Nervosa , Humanos , Imageamento por Ressonância Magnética/métodos , Encéfalo/diagnóstico por imagem , Encéfalo/fisiologia , Análise por Conglomerados , Rede Nervosa/diagnóstico por imagem , Rede Nervosa/fisiologia , Processamento de Imagem Assistida por Computador/métodos , Mapeamento Encefálico/métodos , Redes Neurais de Computação , Conectoma/métodos , Modelos NeurológicosRESUMO
OBJECTIVE: To investigate the genetic prognostic factors for the recurrence of non-muscle invasive bladder cancer. MATERIALS AND METHODS: The patients underwent transurethral resection of bladder tumor and received bacillus Calmette-Guérin (BCG) or epirubicin. Next-generation sequencing was performed and alterations of genes, pathways, and tumor mutation burden were recorded. Associations between these clinicopathological and genetic variants were estimated, and prognostic factor identified. RESULTS: A total of 58 cases were included in our study, and 46 patients underwent treatment with BCG. FGFR3 was the most frequently altered gene (48%), and more commonly detected in intermediate-risk patients. Univariate Cox analysis demonstrated that 10 genes were significantly correlated with BCG failure, while NEB, FGFR1 and SDHC were independent recurrence predictors. Besides, epigenetic-related gene pathway mutations were negatively correlated with recurrence (hazard ratio: 0.198, P = 0.023). DNA damage response and repair gene alterations were positively correlated with tumor burden, while altered TP53 was most frequent among these genes and significant correlated with high tumor burden. CONCLUSION: BCG instillation significantly reduced the rate of recurrence compared with epirubicin in this population. Potential biomarkers and therapeutic targets were found with the help of next-generation sequencing; correlations between DDR genes alterations and high tumor mutation burden were also demonstrated.
RESUMO
This paper is concerned with the recursive state estimation (RSE) problem under minimum mean-square error sense for a class of nonlinear complex networks (CNs) with uncertain inner coupling, random link failures and packet disorders. Firstly, a set of random variables obeying the Bernoulli distribution is adopted to characterize whether there are connections between different network units, i.e., there is no random link failure when the random variable is equal to 1, otherwise the random link failure occurs. In addition, the inner coupling strength is assumed to be varying within a given interval and the phenomenon of packet disorders caused by the random transmission delay (RTD) is also taken into account. In our study, the nonlinearity satisfies the continuously differentiable condition, which can be linearized by resorting to the Taylor expansion. The focus of the addressed RSE problem is on the design of an RSE approach in the mean-square error sense such that, for all uncertain inner coupling, random link failures and packet disorders, a suboptimal upper bound of the state estimation error covariance is obtained and minimized by parameterizing the state estimator gain with explicit expression form. Furthermore, a sufficient condition with respect to the uniform boundedness of state estimation error in mean-square sense is elaborated. Finally, a numerical experiment is introduced to demonstrate the validity of the presented RSE approach.
RESUMO
To compare the clinical feasibility and oncological outcome of different surgical techniques for inguinal lymphadenectomy (ILND) in patients suffering from penile cancer. This study included data from 109 cN0-2 patients diagnosed with penile cancer who received ILND. 80 laparoscopic ILND were performed on 40 patients, while 138 open surgeries were performed on 69 patients. Perioperative complications and prognosis were compared between different surgical techniques. Compared with the open surgery group, the laparoscopy group had a shorter hospital stay (8.88 ± 7.86 days vs. 13.94 ± 10.09 days, P = 0.004), and a lower wound healing delay rate (8.75% vs. 22.46%, P = 0.017), but also had longer drainage time (10.91 ± 9.66 vs. 8.70 ± 4.62, P = 0.002). There were no significant differences in terms of other intraoperative parameters, complications, and survival between open and laparoscopic group. Compared with saphenous vein ligated subgroup, preserved subgroup showed no significant reducing of complication rate. There was no significant difference among complication between different open surgery subgroup. Immediate ILND showed no prognostic advantage over delayed ILND regardless of clinical lymph node status. Compared with open surgery, the minimally invasive ILND technique has similar oncological efficiency and a lower complication rate. Saphenous vein preservation has limited value in reducing complications. Delayed lymphadenectomy might be a more reasonable option for ILND.
Assuntos
Laparoscopia/métodos , Excisão de Linfonodo/métodos , Neoplasias Penianas/cirurgia , Humanos , Masculino , PrognósticoRESUMO
Background: Poly(ADP-ribose) polymerase (PARP) inhibitors have breakthrough designations for metastatic castration-resistant prostate cancer (mCRPC). We performed a meta-analysis of current clinical trials to evaluate the efficacy of PARP inhibitors in mCRPC patients based on their genetic status. Methods: On August 2020, PubMed, Scopus, Embase, Cochrane Central Register of Controlled Trials, and Web of Science were searched for phase II/III clinical studies on PARP inhibitors in mCRPC patients. Data were extracted independently by two investigators and analyzed using Review Manager software version 5.3. Primary endpoints included overall response rate (ORR) and progression-free survival (PFS). Results: Nine clinical trials were identified and analyzed for the clinical benefit of PARP inhibitors in mCRPC patients (n = 1,219). Pooled analyses demonstrated that PARP inhibitors could provide a significant improvement of ORR and PFS in patients with homologous recombination deficiency (HRD) when compared with non-HRD patients. Within the HRD subgroup, BRCA mutation patients achieved significantly higher ORR [odds ratio (OR): 9.97, 95% confidence interval (CI): 6.08-16.35] and PFS rates at 12 months (OR: 3.23, 95% CI: 1.71-6.10) when compared with BRCA wild-type patients. Furthermore, patients harboring HRD without BRCA mutations have a higher objective response after PARP inhibitor treatment compared with non-HRD patients. Conclusion: PARP inhibitor is an effective treatment option for mCRPC patients with mutations in genes related to the HR DNA repair pathway when compared with non-HRD patients. In addition to BRCA mutations, other HRD-related gene aberrations may also be used as novel biomarkers to predict the efficacy of PARP inhibitors.
RESUMO
OBJECTIVE: To evaluate the prognostic factors of penile cancer and the utility of prognostic models. METHODS: We analyzed postoperatively collected data of 311 patients diagnosed with penile cancer. Survival analysis (Kaplan-Meier and cox regression methods) was performed on this cohort. The c-index was used to determine the predictive accuracies of potential prognostic factors. The accuracies of four prognostic models were also evaluated, which were AJCC prognostic stage group for three recent editions, and four nomograms constructed by the Surveillance, Epidemiology, and End Results program (SEER). Two novel nomograms using our data were created and AUC of 2-year survival were determined to compare existing and newly established models. RESULTS: Tumor site, T and N stages, nuclear grade and lymph vascular invasion (LVI) significantly influenced prognosis. The 8th T and N stages had better c-indexes than former editions, while no improvement was seen in the 8thAJCC stage group. 6th AJCC+grade nomogram had a higher c-index than other three nomograms (SEER+grade, 6th TNM+grade, and 6th T1-3N0-3+grade nomograms; c-index: 0.831 vs 0.738, 0.792 and 0.781). New nomogram 1 included the 8th T and N stages, tumor site, nuclear grade, and LVI, with a c-index of 0.870. Novel nomogram 2 replaced the T and N stages with the AJCC stage group, which had a lower c-index of 0.855. The order of prediction accuracy of 2-year survival in the old and new models is consistent with the c-index results. CONCLUSION: Tumor site, stages, grade, and LVI play important roles in predicting survival of penile cancer. The 8th stages have better predictive accuracy than former editions. We proposed two models with better predictive accuracy than former models; specifically, nomogram 1 may be a more precise and convenient tool for predicting penile cancer outcomes.
RESUMO
High-affinity K(+) transporters play an important role in K(+) absorption of plants. We isolated a HAK gene from Aeluropus littoralis, a graminaceous halophyte. The amino acid sequence of AlHAK showed high homology with HAK transporters obtained from Oryza sativa (82%) and Hordeum vulgare (82%). When expressed in Saccharomyces cereviae WDelta3, AlHAK performed high-affinity K(+) uptake with a K(m) value of 8 muM, and the growth of transformants was dramatically inhibited by 150 mM Rb(+) and 150 mM Cs(+) but less affected by 300 mM Na(+). AlHAK may thus improve the capacity of plants to maintain a high cytosolic K(+)/Na(+) ratio at high salinity.