Software-Automated Implant Detection for Intraoperative 3D Imaging—First Clinical Evaluation on 214 Data Sets

Previous studies have demonstrated a frequent occurrence of screw/K-wire malpositioning during surgical fracture treatment under 2D fluoroscopy and a correspondingly high revision rate as a result of using intraoperative 3D imaging. In order to facilitate and accelerate the diagnosis of implant malpositioning in 3D data sets, this study investigates two versions of an implant detection software for mobile 3D C-arms in terms of their detection performance based on comparison with manual evaluation. The 3D data sets of patients who had received surgical fracture treatment at five anatomical regions were extracted from the research database. First, manual evaluation of the data sets was performed, and the number of implanted implants was assessed. For 25 data sets, the time required by four investigators to adjust each implant was monitored. Subsequently, the evaluation was performed using both software versions based on the following detection parameters: true-positive-rate, false-negative-rate, false-detection-rate and positive predictive value. Furthermore, the causes of false positive and false negative detected implants depending on the anatomical region were investigated. Two hundred fourteen data sets with overall 1767 implants were included. The detection parameters were significantly improved (p<.001) from version 1 to version 2 of the implant detection software. Automatic evaluation required an average of 4.1±0.4 s while manual evaluation was completed in 136.15±72.9 s (p<.001), with a statistically significant difference between experienced and inexperienced users (p=.005). In summary, version 2 of the implant detection software achieved significantly better results. The time saved by using the software could contribute to optimizing the intraoperative workflow.     

Changes in Alcohol Consumption,Eating Behaviors,and Body Weight during Quarantine Measures: Analysis of the CoCo-Fakt Study

IntroductionPublic health measures enacted to reduce COVID-19 transmission have affected individuals'' lifestyles, mental health, and psychological well-being. To date, little is known how stay-at-home orders have influenced the eating behaviors, weight development, and alcohol consumption of quarantined persons. The CoCo-Fakt cohort study analyzed these parameters and their association with psychological distress and coping strategies.MethodsAn online survey was conducted of all persons who tested positive for SARS-CoV-2 (infected persons [IP]) between December 12, 2020, and January 6, 2021, as well as their close contacts (contact persons [CP]) registered by the public health department of Cologne. 8,075 of 33,699 individuals were included in the analysis. In addition to demographic data, psychological distress, and coping strategies, information on changes in body weight, eating, and drinking behaviors was collected.ResultsIP lost 1.2 ± 4.4 kg during the quarantine period, and CP gained 1.6 ± 4.1 kg. The reasons given by IP for weight change were mainly loss of taste and feeling sick, whereas CP were more likely than IP to eat out of boredom. Higher psychological burden and lower coping strategies were associated with both weight gain and loss. Of the 30.8% of participants who changed their alcohol consumption during the quarantine period, CP in particular drank more alcohol (IP 15.2%; CP 47.7%). Significantly less alcohol was consumed by individuals with higher coping scores.ConclusionIn this short but psychologically stressful period of stay-at-home orders, changes in eating and drinking behavior as well as weight development are evident, mainly in high-risk contacts. To avoid possible long-term sequelae, health authorities should take these findings into account during the quarantine period; in particular, general practitioners should consider these findings during follow-up.     

Limited Effects of SARS-CoV-2 Pandemic-related Lockdowns and Reduced Population Mobility on Preterm Birth Rates: A Secondary Analysis of Bavarian Obstetric Quality Parameters from 2010 to 2020

Introduction International studies on preterm birth rates during COVID-19 lockdowns report different results. This study examines preterm birth rates during lockdown periods and the impact of the mobility changes of the population in Bavaria, Germany. Material and Methods This is a secondary analysis of centrally collected data on preterm births in Bavaria from 2010 to 2020. Preterm births (< 37 weeks) in singleton and twin pregnancies during two lockdowns were compared with corresponding periods in 2010 – 2019. Fisherʼs exact test was used to compare raw prevalence between groups. Potential effects of two fixed lockdown periods and of variable changes in population mobility on preterm birth rates in 2020 were examined using additive logistic regression models, adjusting for long-term and seasonal trends. Results Unadjusted preterm birth rates in 2020 were significantly lower for singleton pregnancies during the two lockdown periods (Lockdown 1: 5.71% vs. 6.41%; OR 0.88; p < 0.001; Lockdown 2: 5.71% vs. 6.60%; OR = 0.86; p < 0.001). However, these effects could not be confirmed after adjusting for long-term trends (Lockdown 1: adj. OR = 0.99; p = 0.73; Lockdown 2: adj. OR = 0.96; p = 0.24). For twin pregnancies, differences during lockdown were less marked (Lockdown 1: 52.99% vs. 56.26%; OR = 0.88; p = 0.15; Lockdown 2: 58.06% vs. 58.91%; OR = 0.97; p = 0.70). Reduced population mobility had no significant impact on preterm birth rates in singleton pregnancies (p = 0.14) but did have an impact on twin pregnancies (p = 0.02). Conclusions Reduced preterm birth rates during both lockdown periods in 2020 were observed for singleton and twin pregnancies. However, these effects are reduced when adjusting for long-term and seasonal trends. Reduced population mobility was associated with lower preterm birth rates in twin pregnancies. Key words: COVID-19, SARS-CoV-2, preterm birth, lockdown measures, preterm delivery     

Improving GWAS discovery and genomic prediction accuracy in biobank data

Genetically informed, deep-phenotyped biobanks are an important research resource and it is imperative that the most powerful, versatile, and efficient analysis approaches are used. Here, we apply our recently developed Bayesian grouped mixture of regressions model (GMRM) in the UK and Estonian Biobanks and obtain the highest genomic prediction accuracy reported to date across 21 heritable traits. When compared to other approaches, GMRM accuracy was greater than annotation prediction models run in the LDAK or LDPred-funct software by 15% (SE 7%) and 14% (SE 2%), respectively, and was 18% (SE 3%) greater than a baseline BayesR model without single-nucleotide polymorphism (SNP) markers grouped into minor allele frequency–linkage disequilibrium (MAF-LD) annotation categories. For height, the prediction accuracy R² was 47% in a UK Biobank holdout sample, which was 76% of the estimated hSNP2. We then extend our GMRM prediction model to provide mixed-linear model association (MLMA) SNP marker estimates for genome-wide association (GWAS) discovery, which increased the independent loci detected to 16,162 in unrelated UK Biobank individuals, compared to 10,550 from BoltLMM and 10,095 from Regenie, a 62 and 65% increase, respectively. The average χ2 value of the leading markers increased by 15.24 (SE 0.41) for every 1% increase in prediction accuracy gained over a baseline BayesR model across the traits. Thus, we show that modeling genetic associations accounting for MAF and LD differences among SNP markers, and incorporating prior knowledge of genomic function, is important for both genomic prediction and discovery in large-scale individual-level studies.

As biobank datasets increase in size, it is important to understand the factors limiting the prediction of phenotype from genotype. Alongside others, we have recently shown that genomic prediction accuracy can be improved through the use of random-effects models that incorporate prior knowledge of genomic annotations and allow for differences in the variance explained by single-nucleotide polymorphism (SNP) markers, depending upon their linkage disequilibrium (LD) and their minor allele frequency (MAF) (1–8). These improvements in prediction accuracy should also translate into greater genome-wide association study (GWAS) discovery power. Mixed-linear models of association (MLMA) are commonly applied in GWASs in a two-step approach, where a random-effects model is first used to estimate leave-one-chromosome-out (LOCO) genetic values, and these are then used in a second marginal regression coefficient estimation step. Theory suggests that the test statistics obtained in the MLMA second step depend upon the accuracy of the LOCO genomic predictors produced from the first step. Current MLMA implementations use a blocked ridge regression model (9), a Restricted Maximum Likelihood (REML) genomic relationship model (10), or a Bayesian spike-and-slab model (11) within the first step.Here, we improve the computational implementation of our recently developed Bayesian grouped mixture of regressions model (GMRM), which estimates genetic marker effects jointly, but with independent marker inclusion probabilities and independent hSNP2 parameters across LD, MAF, and functional annotation groups (Materials and Methods). This allows us to apply the model to 21 traits in the UK Biobank to test for prediction accuracy improvements over existing approaches. We then extend the model to provide MLMA SNP marker association estimates to test whether improved prediction accuracy translates to improved GWAS discovery compared to existing MLMA approaches.     

Suramin induces and enhances apoptosis in a model of hyperoxia-induced oligodendrocyte injury

Recent evidence suggests oxygen as a powerful trigger for cell death in the immature white matter, leading to periventricular leukomalacia (PVL) as a cause of adverse neurological outcome in survivors of preterm birth. This oligodendrocyte (OL) death is associated with oxidative stress, upregulation of apoptotic signaling factors (i.e., Fas, caspase-3) and decreased amounts of neurotrophins. In search of neuroprotective strategies we investigated whether the polysulfonated urea derivative suramin, recently identified as a potent inhibitor of Fas signaling, affords neuroprotection in an in vitro model of hyperoxia-induced injury to immature oligodendrocytes. Immature OLs (OLN-93) were subjected to 80% hyperoxia (48 h) in the presence or absence of suramin (0, 30, 60, 120 microM). Cell death was assessed by flow cytometry (Annexin V, caspase-3 activity assay) and immunohistochemistry for activated caspase-3. Immunoblotting for the death receptor Fas, cleaved caspase-8 and the phosphorylated isoform of the serine-threonin kinase Akt (pAkt) was performed. Suramin lead to OL apoptosis and potentiated hyperoxia-induced injury in a dose-dependent manner. Immunoblotting revealed increased Fas and caspase-8 expression by suramin treatment. This effect was significantly enhanced when suramin was combined with hyperoxia. Furthermore, pAkt levels decreased following suramin exposure, indicating interference with neurotrophin-dependent growth factor signaling. These data indicate that suramin causes apoptotic cell death and aggravates hyperoxia-induced cell death in immature OLs. Its mechanism of action includes an increase of previously described hyperoxia-induced expression of pro-apoptotic factors and deprivation of growth factor dependent signaling components.     

Two types of mitochondrial crystals in diseased human skeletal muscle fibers

Mitochondrial crystalline inclusions, frequently found in mitochondrial myopathies, were analyzed by crystallographic techniques and computer-aided image processing. It could be shown that these structures were real crystals. There are two distinct types of crystal, which can be distinguished by shape, size, and pattern. So-called type I crystals are usually present in the intracristal space, whereas the type II crystals are preferentially located in the intermembrane space between outer and inner mitochondrial membranes. The unit cell dimensions were found to be 38 x 34 x 8 nm for the type I crystals and 20 x 17 x 8 nm for the type II crystals. These results strongly suggest that the crystals are composed of macromolecules, presumably proteins. Arguments are presented that indicate that type I crystals occur only in type 1 muscle fibers and type II crystals in type 2 muscle fibers.