Tag: statistics

J Neurooncol. 2023 Feb 25. doi: 10.1007/s11060-022-04233-y. Online ahead of print.

ABSTRACT

PURPOSE: To comprehensively investigate prognostic factors, including clinical and molecular factors and treatment modalities, in adult glioma patients with leptomeningeal metastases (LM).

METHODS: Total 226 patients with LM (from 2001 to 2021 among 1495 grade 2 to 4 glioma patients, 88.5% of LM patients being IDH-wildtype) with complete information on IDH mutation, 1p/19q codeletion, and MGMT promoter methylation status were enrolled. Predictors of overall survival (OS) of entire patients were determined by time-dependent Cox analysis, including clinical, molecular, and treatment data. Subgroup analyses were performed for patients with LM at initial diagnosis and LM diagnosed at recurrence (herein, initial and recurrent LM). Identical analyses were performed in IDH-wildtype glioblastoma patients.

RESULTS: Median OS was 17.0 (IQR 9.7-67.1) months, with shorter median OS in initial LM than recurrent LM patients (12.2 vs 20.6 months, P < 0.001). In entire patients, chemotherapy and antiangiogenic therapy were predictors of longer OS, while male sex and initial LM were predictors of shorter OS. In initial LM, higher KPS, chemotherapy, and antiangiogenic therapy were predictors of longer OS, while male sex was a predictor of shorter OS. In recurrent LM, chemotherapy and longer interval between initial glioma and LM diagnoses were predictors of longer OS, while male sex was a predictor of shorter OS. A similar trend was observed in IDH-wildtype glioblastoma.

CONCLUSION: Active chemotherapy and antiangiogenic therapy demonstrated survival benefit in glioma patients with LM. There is consistent female survival advantage, whereas longer interval between initial glioma diagnosis and LM development suggests longer OS in recurrent LM.

PMID:36841906 | DOI:10.1007/s11060-022-04233-y

Ann Biomed Eng. 2023 Feb 25. doi: 10.1007/s10439-023-03156-7. Online ahead of print.

ABSTRACT

The use of ovine animal models in the study of injury biomechanics and modeling is increasing, due to their favorable size and other physiological characteristics. Along with this increase, there has also been increased interest in the development of in silico ovine models for computational studies to compliment physical experiments. However, there remains a gap in the literature characterizing the morphological and mechanical characteristics of ovine ribs. The objective of this study therefore is to report anatomical and mechanical properties of the ovine ribs using microtomography (micro-CT) and two types of mechanical testing (quasi-static bending and dynamic tension). Using microtomography, young ovine rib samples obtained from a local abattoir were cut into approximately fourteen 38 mm sections and scanned. From these scans, the cortical bone thickness and cross-sectional area were measured, and the moment of inertia was calculated to enhance the mechanical testing data. Based on a standard least squares statistical model, the cortical bone thickness varied depending on the region of the cross-section and the position along the length of the rib (p < 0.05), whereas the cross-sectional area remained consistent (p > 0.05). Quasi-static three-point bend testing was completed on ovine rib samples, and the resulting force-displacement data was analyzed to obtain the stiffness (44.67 ± 17.65 N/mm), maximum load (170.54 ± 48.28 N) and displacement at maximum load (7.19 ± 2.75 mm), yield load (167.81 ± 48.12 N) and displacement at yield (6.10 ± 2.25 mm), and the failure load (110.90 ± 39.30 N) and displacement at failure (18.43 ± 2.10 mm). The resulting properties were not significantly affected by the rib (p > 0.05), but by the animal they originated from (p < 0.05). For the dynamic testing, samples were cut into coupons and tested in tension with an average strain rate of 18.9 strain/sec. The resulting dynamic testing properties of elastic modulus (5.16 ± 2.03 GPa), failure stress (63.29 ± 14.02 MPa), and failure strain (0.0201 ± 0.0052) did not vary based on loading rate (p > 0.05).

PMID:36841890 | DOI:10.1007/s10439-023-03156-7

Sci Data. 2023 Feb 25;10(1):113. doi: 10.1038/s41597-023-02013-5.

ABSTRACT

Measuring and monitoring the spatial distribution of liveability is crucial to ensure that implemented urban and transport planning decisions support health and wellbeing. Spatial liveability indicators can be used to ensure these decisions are effective, equitable and tracked across time. The 2018 Australian National Liveability Study datasets comprise a suite of policy-relevant health-related spatial indicators of local neighbourhood liveability and amenity access estimated for residential address points and administrative areas across Australia’s 21 most populous cities. The indicators and measures encompass access to community and health services, social infrastructure, employment, food, housing, public open space, transportation, walkability and overall liveability. This national ‘baseline’ liveability indicators dataset for residential address points and areas can be further linked with surveys containing geocoded participant locations, as well as Census data for areas from the Australian Statistical Geography Standard. The datasets will be of interest to planners, policy makers and researchers interested in modelling and mapping the spatial distribution of urban environmental exposures and their relationship with health and other outcomes.

PMID:36841876 | DOI:10.1038/s41597-023-02013-5

Sci Rep. 2023 Feb 25;13(1):3269. doi: 10.1038/s41598-023-30365-w.

ABSTRACT

Air pollution due to air contamination by gases, liquids, and solid particles in suspension, is a great environmental and public health concern nowadays. An important type of air pollution is particulate matter with a diameter of 10 microns or less ([Formula: see text]) because one of the determining factors that affect human health is the size of particles in the atmosphere due to the degree of permanence and penetration they have in the respiratory system. Therefore, it is extremely interesting to monitor and understand the behavior of [Formula: see text] concentrations so that they do not exceed the established critical levels. In this work, we will study the [Formula: see text] concentrations in all available monitoring stations in the Brazilian state of Minas Gerais. To better understand its behavior, we will provide a spatio-temporal visualization of the [Formula: see text] concentrations. Besides the descriptive and visualization analysis, we consider six standard and advanced time series models that will be used to fit and forecast [Formula: see text] concentrations, with application to three locations, one in Belo Horizonte, the Minas Gerais state capital, and the monitoring stations with the lowest and highest average [Formula: see text] concentration levels.

PMID:36841859 | DOI:10.1038/s41598-023-30365-w

Sci Rep. 2023 Feb 25;13(1):3292. doi: 10.1038/s41598-023-29132-8.

ABSTRACT

Recent advances in technology have led to an explosion of data in virtually all domains of our lives. Modern biomedical devices can acquire a large number of physical readings from patients. Often, these readings are stored in the form of time series data. Such time series data can form the basis for important research to advance healthcare and well being. Due to several considerations including data size, patient privacy, etc., the original, full data may not be available to secondary parties or researchers. Instead, suppose that a subset of the data is made available. A fast and reliable record linkage algorithm enables us to accurately match patient records in the original and subset databases while maintaining privacy. The problem of record linkage when the attributes include time series has not been studied much in the literature. We introduce two main contributions in this paper. First, we propose a novel, very efficient, and scalable record linkage algorithm that is employed on time series data. This algorithm is 400× faster than the previous work. Second, we introduce a privacy preserving framework that enables health institutions to safely release their raw time series records to researchers with bare minimum amount of identifying information.

PMID:36841850 | DOI:10.1038/s41598-023-29132-8

Sci Rep. 2023 Feb 25;13(1):3288. doi: 10.1038/s41598-023-29495-y.

ABSTRACT

A kernel-based quantum classifier is the most practical and influential quantum machine learning technique for the hyper-linear classification of complex data. We propose a Variational Quantum Approximate Support Vector Machine (VQASVM) algorithm that demonstrates empirical sub-quadratic run-time complexity with quantum operations feasible even in NISQ computers. We experimented our algorithm with toy example dataset on cloud-based NISQ machines as a proof of concept. We also numerically investigated its performance on the standard Iris flower and MNIST datasets to confirm the practicality and scalability.

PMID:36841841 | DOI:10.1038/s41598-023-29495-y

Sci Rep. 2023 Feb 25;13(1):3276. doi: 10.1038/s41598-022-27276-7.

ABSTRACT

The United Nations projects that one in every six people will be over the age of 65 by the year 2050. With a rapidly aging population, the risk of Alzheimer’s disease (AD) becomes a major concern. AD is a multifactorial disease that involves neurodegeneration in the brain with mild dementia and deficits in memory and other cognitive domains. Additionally, it has been established that individuals with Human Immunodeficiency Virus-1 (HIV-1) experience a 5 to 10-year accelerated aging and an increased risk of developing HIV-associated neurocognitive disorders (HAND). Despite a significant amount of clinical evidence pointing towards a potential overlap between neuropathogenic processes in HAND and AD, the underlying epigenetic link between these two diseases is mostly unknown. This study is focused on identifying differentially expressed genes observed in both AD and HAND using linear regression models and a more robust significance analysis of microarray. The results established that the dysregulated type 1 and 2 interferon pathways observed in both AD and HAND contribute to the similar pathologies of these diseases within the brain. The current study identifies the important roles of interferon pathways in AD and HAND, a relationship that may be useful for earlier detection in the future.

PMID:36841839 | DOI:10.1038/s41598-022-27276-7

Nat Commun. 2023 Feb 25;14(1):1075. doi: 10.1038/s41467-023-36752-1.

ABSTRACT

Endosomal escape and subsequent cytosolic delivery of small interfering RNA (siRNA) therapeutics is believed to be highly inefficient. Since it has not been possible to quantify cytosolic amounts of delivered siRNA at therapeutic doses, determining delivery bottlenecks and total efficiency has been difficult. Here, we present a confocal microscopy-based method to quantify cytosolic delivery of fluorescently labeled siRNA during lipid-mediated delivery. This method enables detection and quantification of sub-nanomolar cytosolic siRNA release amounts from individual release events with measures of quantitation confidence for each event. Single-cell kinetics of siRNA-mediated knockdown in cells expressing destabilized eGFP unveiled a dose-response relationship with respect to knockdown induction, depth and duration in the range from several hundred to thousands of cytosolic siRNA molecules. Accurate quantification of cytosolic siRNA, and the establishment of the intracellular dose-response relationships, will aid the development and characterization of novel delivery strategies for nucleic acid therapeutics.

PMID:36841822 | DOI:10.1038/s41467-023-36752-1

Adv Simul (Lond). 2023 Feb 25;8(1):7. doi: 10.1186/s41077-023-00244-5.

ABSTRACT

Neonatal deaths are a major contributor to global under-5-year-old mortality. Training birth attendants can improve perinatal outcomes, but skills may fade over time. In this pilot study, we assessed skill decay of nursing students after remote video versus in-person resuscitation training in a low-resource setting. Filipino nursing students (n = 49) underwent traditional, in-person simulation-based Helping Babies Breathe (HBB) training in Mindanao, Philippines. Participants were then assigned to receive refresher training at 2-month intervals either in-person or via tele-simulation beginning at 2 months, 4 months, or 6 months after initial training. A knowledge examination and practical examination, also known as objective structured clinical examination B in the HBB curriculum, were administered before retraining to assess knowledge and skill retention at time of scheduled follow-up. Time to initiation of bag-mask ventilation (BMV) in seconds during simulated birth asphyxia was the primary outcome. Skill decay was evident at first follow-up, with average time to BMV increasing from 56.9 (range 15-87) s at initial post-training to 83.8 (range 32-128) s at 2 months and 90.2 (range 51-180) s at 4 months. At second follow-up of the 2-month group, students showed improved pre-training time to BMV (average 70.4; range 46-97 s). No statistical difference was observed between in-person and video-trained students in time to BMV. Because of COVID-19 restrictions, the 6-month follow-up was not completed. We conclude that remote video refresher training is a reasonable alternative to traditional in-person HBB training. Our study also suggests that refreshers may be needed more frequently than every 2 months to mitigate skill decay. Additional studies are necessary to assess the longitudinal impact of tele-simulation on clinical outcomes.

PMID:36841812 | DOI:10.1186/s41077-023-00244-5

Oncogenesis. 2023 Feb 25;12(1):10. doi: 10.1038/s41389-023-00456-4.

ABSTRACT

Head and neck paragangliomas (HNPGLs), rare chemoresistant tumors curable only with surgery, are strongly influenced by genetic predisposition, hence patients and relatives require lifetime follow-up with MRI and/or PET-CT because of de novo disease risk. This entails exposure to electromagnetic/ionizing radiation, costs, and organizational challenges, because patients and relatives are scattered far from reference centers. Simplified first-line screening strategies are needed. We employed flow injection analysis tandem mass spectrometry, as used in newborn metabolic screening, to compare the plasma metabolic profile of HNPGL patients (59 samples, 56 cases) and healthy controls (24 samples, 24 cases). Principal Component Analysis (PCA) and Partial Least Discriminant Analysis (PLS-DA) highlighted a distinctive HNPGL signature, likely reflecting the anaplerotic conversion of the TCA cycle to glutaminolysis and catabolism of branched amino acids, DNA damage and deoxyadenosine (dAdo) accumulation, impairment of fatty acid oxidation, switch towards the Warburg effect and proinflammatory lysophosphatidylcholines (LPCs) signaling. Statistical analysis of the metabolites that most impacted on PLS-DA was extended to 10 acoustic neuroma and 2 cholesteatoma patients, confirming significant differences relative to the HNPGL plasma metabolomic profile. The best confusion matrix from the ROC curve built on 2 metabolites, dAdo and C26:0-LPC, provided specificity of 94.29% and sensitivity of 89.29%, with positive and negative predictive values of 96.2% and 84.6%, respectively. Analysis of dAdo and C26:0-LPC levels in dried venous and capillary blood confirmed that dAdo, likely deriving from 2′-deoxy-ATP accumulated in HNPGL cells following endogenous genotoxic damage, efficiently discriminated HNPGL patients from healthy controls and acoustic neuroma/cholesteatoma patients on easily manageable dried blood spots.

PMID:36841802 | DOI:10.1038/s41389-023-00456-4