Category: Statistics

J Biophotonics. 2022 Jan 25:e202100370. doi: 10.1002/jbio.202100370. Online ahead of print.

ABSTRACT

Recently, convolutional neural networks (CNNs) have been widely utilized for breast cancer histopathology image classification. Besides, research works have also convinced that deep high-order statistic models obviously outperform corresponding first-order counterparts in vision tasks. Inspired by this, we attempt to explore global deep high-order statistics to distinguish breast cancer histopathology images. To further boost the classification performance, we also integrate asymmetric convolution into the second-order network and propose a novel second-order asymmetric convolution network (SoACNet). SoACNet adopts a series of asymmetric convolution blocks to replace each stand square-kernel convolutional layer of the backbone architecture, followed by a global covariance pooling to compute second-order statistics of deep features, leading to a more robust representation of histopathology images. Extensive experiments on the public BreakHis dataset demonstrate the effectiveness of SoACNet for breast cancer histopathology image classification, which achieves competitive performance with the state-of-the-arts. This article is protected by copyright. All rights reserved.

PMID:35076187 | DOI:10.1002/jbio.202100370

Brain Behav. 2022 Jan 25:e02077. doi: 10.1002/brb3.2077. Online ahead of print.

ABSTRACT

BACKGROUND: Passive measures collected using smartphones have been suggested to represent efficient proxies for depression severity, but the performance of such measures across diagnoses has not been studied.

METHODS: We enrolled a cohort of 45 individuals (11 with major depressive disorder, 11 with bipolar disorder, 11 with schizophrenia or schizoaffective disorder, and 12 individuals with no axis I psychiatric disorder). During the 8-week study period, participants were evaluated with a rater-administered Montgomery-Åsberg Depression Rating Scale (MADRS) biweekly, completed self-report PHQ-8 measures weekly on their smartphone, and consented to collection of smartphone-based GPS and accelerometer data in order to learn about their behaviors. We utilized linear mixed models to predict depression severity on the basis of phone-based PHQ-8 and passive measures.

RESULTS: Among the 45 individuals, 38 (84%) completed the 8-week study. The average root-mean-squared error (RMSE) in predicting the MADRS score (scale 0-60) was 4.72 using passive data alone, 4.27 using self-report measures alone, and 4.30 using both.

CONCLUSIONS: While passive measures did not improve MADRS score prediction in our cross-disorder study, they may capture behavioral phenotypes that cannot be measured objectively, granularly, or over long-term via self-report.

PMID:35076166 | DOI:10.1002/brb3.2077

Head Neck. 2022 Jan 25. doi: 10.1002/hed.26987. Online ahead of print.

ABSTRACT

INTRODUCTION: This study aimed to systematically elucidate the metastatic patterns and their corresponding survival of each thyroid cancer subtype at time of diagnosis.

METHODS: We accessed the Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2018 to search for primary thyroid cancers with DM at presentation (M1).

RESULTS: We included 2787 M1 thyroid cancers for statistical analyses and the incidence of DM at presentation was 2.4%. Lung was the most common metastatic site for anaplastic thyroid carcinoma (ATC), poorly differentiated thyroid carcinoma (PDTC), papillary thyroid carcinoma (PTC), and oncocytic (Hurthle) cell carcinoma (HCC) whereas bone is the favorable disseminated site of follicular thyroid carcinoma (FTC) and medullary thyroid carcinoma (MTC). Patients with multi-organ metastases had the worst survival whereas bone metastases were associated with a favorable outcome (p < 0.001).

CONCLUSION: There are significant differences in DM patterns of thyroid cancer subtypes and their corresponding survival.

PMID:35076146 | DOI:10.1002/hed.26987

Headache. 2022 Jan 25. doi: 10.1111/head.14257. Online ahead of print.

ABSTRACT

OBJECTIVE: The purpose of this study was to propose a definition of “wearing off” at the individual patient-level and determine the percentage of patients with migraine who experience “wearing off” of efficacy of galcanezumab at the end of a treatment cycle using this predefined threshold.

BACKGROUND: Anecdotal reports suggest that some patients may experience “wearing off” of efficacy during the last week of their calcitonin gene-related peptide monoclonal antibody treatment cycle. A previous post hoc analysis of galcanezumab demonstrated consistent efficacy at each week throughout all monthly dosing intervals at the population-level, but “wearing off” has not been assessed at the individual patient-level.

METHODS: Post hoc analyses of clinical trial data from four galcanezumab phase III, randomized, placebo-controlled studies in a total of 2680 patients with high-frequency episodic migraine (EVOLVE-1, EVOLVE-2, and CONQUER studies) or chronic migraine (CM; REGAIN and CONQUER studies) were conducted. “Wearing off” was defined as an increase of greater than or equal to 2 weekly migraine headache days in the last week of the treatment cycle compared to the second week for at least 2 months. The analyses were conducted (1) in all patients and (2) in patients with a clinically meaningful response to treatment.

RESULTS: The percentage of patients meeting the threshold of “wearing off” was not statistically significantly different among the placebo, galcanezumab 120 mg, and galcanezumab 240 mg treatment groups, both in the total population and in patients with a clinically meaningful response (all ≤9.0%). Although the frequency of “wearing off” in patients with CM and prior preventive failures was numerically greater in the galcanezumab groups (8/89 or 9.0%) compared to placebo (3/95 or 3.2%), these differences were not statistically significant.

CONCLUSIONS: Consistent with previous analyses at the population-level that showed no evidence of decreased efficacy at the end of a treatment cycle, rates of individual patients meeting the threshold of “wearing off” in this analysis were low and similar among placebo, galcanezumab 120 mg, and galcanezumab 240 mg treatment groups.

PMID:35076090 | DOI:10.1111/head.14257

J Public Health Dent. 2022 Jan 25. doi: 10.1111/jphd.12493. Online ahead of print.

ABSTRACT

INTRODUCTION: A multi-disciplinary, community-university-government collaboration resulted in the development of Oral Health for Community Workers, a brief, oral health online e-learning module. The e-learning module was designed to improve oral health literacy among frontline health workers who are members of underserved communities, and to address oral health disparities in safety net medical and behavioral health settings.

METHODS: Community-based participatory methods were used to design and evaluate the e-learning module. Participants took pre-, post-, and 3-month follow-up surveys.

RESULTS: Oral health literacy and confidence in incorporating oral health into practice improved. Satisfaction with the module was high.

DISCUSSION: Oral Health for Community Workers is now sustained as a standard module within Community Health Worker, Peer Support Specialist, and Peer Recovery Coach Certification and continuing education offerings in Michigan.

PMID:35076094 | DOI:10.1111/jphd.12493

Eur J Cardiothorac Surg. 2022 Jan 24:ezac017. doi: 10.1093/ejcts/ezac017. Online ahead of print.

ABSTRACT

OBJECTIVES: The benefit of rapid-deployment aortic valve replacement (RD-AVR) in patients with a bicuspid aortic valve (BAV) is controversial due to aortic root asymmetry and potential increased risk for valve dislocation and paravalvular leak. This study aimed to analyse the outcomes of surgical aortic valve replacement with a rapid-deployment bioprosthesis in patients with a BAV.

METHODS: Between May 2010 and December 2020, all consecutive patients who underwent RD-AVR at the Medical University of Vienna were included in our institutional database. Assessment of preoperative characteristics, operative outcomes, long-term survival and clinical events was performed. The outcomes of patients presenting with a native BAV were compared with a control group of patients with native tricuspid valve (TAV); reoperative aortic valve replacements were excluded.

RESULTS: Out of 816 patients, who underwent RD-AVR at our institution, 107 patients with a BAV, mean age 68 (standard deviation: 8) years, were compared with a control group of 690 patients with a TAV, mean age 74 (standard deviation: 7) years; patients presenting with a BAV were significantly younger than patients with a TAV (P < 0.001). Concomitant procedures were performed in 44 (41.1%) patients in the BAV group and in 339 (49.1%) patients in the TAV group (P = 0.123); surgery of the ascending aorta was necessary in 24 (22.4%) in the BAV group, compared with 29 (4.2%) in the control group (P < 0.001). The 5-year cumulative incidence of moderate-to-severe paravalvular regurgitation in the BAV group was 10.7% [95% confidence interval (CI): 4.2-20.7%] and 3.9% (95% CI: 2.4-6.1%) in the TAV group (P = 0.057). Reoperation with valve explantation due to non-structural valve dysfunction at 5 years was 2.8% (95% CI: 0.5-8.8%) in the BAV group, compared to 1.9% (95% CI: 1.0-3.2%) in the TAV cohort (P = 0.89). The overall long-term survival rate in the BAV group was 92% (95% CI: 81-97%) at 5 years and 88% (95% CI: 73-95%), at 10 years, significantly better compared to the TAV group (log-rank test P = 0.002).

CONCLUSIONS: RD-AVR can be performed in patients with a BAV with convincingly medical outcomes. However, a trend to increased frequency of moderate-severe paravalvular regurgitation was observed at long-term follow-up. Consequently, a different surgical approach, compared to tricuspid valves, with distinctly specific technical- and anatomical considerations and requirements, is recommended.

PMID:35076066 | DOI:10.1093/ejcts/ezac017

Elife. 2022 Jan 25;11:e71332. doi: 10.7554/eLife.71332.

ABSTRACT

Urinary metabolic profiling is a promising powerful tool to reflect dietary intake and can help understand metabolic alterations in response to diet quality. Here, we used ¹H NMR spectroscopy in a multicountry study in European children (1147 children from 6 different cohorts) and identified a common panel of 4 urinary metabolites (hippurate, N-methylnicotinic acid, urea, and sucrose) that was predictive of Mediterranean diet adherence (KIDMED) and ultra-processed food consumption and also had higher capacity in discriminating children’s diet quality than that of established sociodemographic determinants. Further, we showed that the identified metabolite panel also reflected the associations of these diet quality indicators with C-peptide, a stable and accurate marker of insulin resistance and future risk of metabolic disease. This methodology enables objective assessment of dietary patterns in European child populations, complementary to traditional questionary methods, and can be used in future studies to evaluate diet quality. Moreover, this knowledge can provide mechanistic evidence of common biological pathways that characterize healthy and unhealthy dietary patterns, and diet-related molecular alterations that could associate to metabolic disease.

PMID:35076016 | DOI:10.7554/eLife.71332

Eur J Cardiothorac Surg. 2022 Jan 24:ezac020. doi: 10.1093/ejcts/ezac020. Online ahead of print.

ABSTRACT

OBJECTIVES: Prior studies have reported a high mortality and incidence of post-repair pulmonary venous obstruction (PVO) in mixed total anomalous pulmonary venous connection (TAPVC). This study sought to review the surgical outcomes in this entity.

METHODS: A review of 61 patients undergoing surgical repair of mixed TAPVC was conducted. Patients with a single ventricle were excluded. Patients were subdivided into 3 groups according to Chowdhury’s classification. Predictors for death and postoperative PVO were explored by Cox regression model.

RESULTS: This study trended towards an older cohort with a median age of 88 days (interquartile range, 56.5-177). Twelve patients belonged to ‘2 + 2’ type, 40 belonged to ‘3 + 1’ type and the remaining 9 belonged to bizarre type. There were no early death and 7 late deaths. Follow-up was available in 96.7% of the patients after discharge with a median duration of 53 months (range, 1-177). Nineteen patients developed post-repair PVO among whom 2 required reintervention. Patients with preoperative PVO had a 4-fold higher risk (95% confidence interval, 1.36-12.38) of postoperative PVO than those without and were more likely to die (P = 0.009). No statistical difference was observed among the 3 subgroups in terms of mortality (P = 0.058) and postoperative PVO (P = 0.186).

CONCLUSIONS: Preoperative PVO was significantly associated with postoperative PVO. There was no statistical difference in terms of death and postoperative PVO among the 3 subtypes of mixed TAPVC. Mid-term results favoured a complete rechanneling of pulmonary veins in ‘3 + 1’ type.

PMID:35076062 | DOI:10.1093/ejcts/ezac020

J Biomol Struct Dyn. 2022 Jan 25:1-14. doi: 10.1080/07391102.2022.2029571. Online ahead of print.

ABSTRACT

An array of computational approaches DFT/QSAR/POM methods has been used for a better understanding of drug properties regarding 13 inhibitor derivatives containing either P2 cyclopentane P1 carboxylic acid moiety (1-9) or a P1 cyclopropyl acyl sulfonamide (10-13). To further recognize binding interactions and their activity trends, molecular docking studies were carried out with the use of HCV, which can be used to accurately predict the interactions of ligands with the receptor. The QSAR models are developed through the use of Multiple Linear Regression (MLR) together with Principal Component Analysis (PCA) methods. The statistical results indicate the multiple correlation coefficient R² = 0.840, which shows favorable estimation stability, as well as showing a significant correlation between the HCV NS3 protease of the studied compounds and their electron-accepting ability. The POM analysis of the Physico-chemical properties of compounds 1-13, shows that they are bearing (O1, O2) and/or (O1, O2, O3) antiviral pockets, whereby all oxygen atoms are Osp2 and bearing negative charges. Similar to the reference ligand (F9K), the most active compound 10 was bound deeply into the binding cavity of NS3 protease making interactions with the residues Gly137, His57, Ala157, and His528. The anti-hepatitis pharmacophore site is similar to the anti-HIV pharmacophore site.

PMID:35075979 | DOI:10.1080/07391102.2022.2029571

Int J Phytoremediation. 2022 Jan 25:1-12. doi: 10.1080/15226514.2022.2027865. Online ahead of print.

ABSTRACT

Direct discharge of raw domestic sewage enriched with nitrogenous and phosphorous compounds into the water bodies causes eutrophication and other environmental hazards with detrimental impacts on public and ecosystem health. The present study focuses on phycoremediation of gray water with Arthrospira sp. using an innovative hydrophobic ceramic membrane-based photobioreactor system integrated with CO₂ biofixation and biodiesel production, aiming for green technology development. Surfactant and oil-rich gray water collected from the domestic kitchen was used wherein, chloride, sulfate, and surfactant concentrations were statistically optimized using response surface methodology (RSM), considering maximum microalgal growth rate as a response for the design. Ideal concentrations (mg/L) of working parameters were found to be 7.91 (sulfate), 880.49 (chloride), and 144.02 (surfactant), respectively to achieve optimum growth rate of 0.43 g_dwt/L/day. Enhancement of growth rate of targeted microalgae by 150% with suitable CO₂ (19.5%) supply and illumination in the photobioreactor affirms its efficient operation. Additionally, harvested microalgal biomass obtained from the process showed a biodiesel content of around 5.33% (dry weight). The microalgal treatment enabled about 96.82, 87.5, and 99.8% reductions in BOD, COD, and TOC, respectively, indicating the potential of the process in pollutant assimilation and recycling of such wastewater along with value-added product generation.

PMID:35075966 | DOI:10.1080/15226514.2022.2027865