Graphical Methods for Data Analysis

Chambers, John M.; Cleveland, William S.; Kleiner, Beat; Tukey, Paul A.

doi:10.1201/9781351072304

Cited by 420 publications

(397 citation statements)

References 0 publications

Supporting

Mentioning

372

Contrasting

Unclassified

Order By: Relevance

“…The locally weighted scatterplot smoothing (Lowess) technique was used to describe the relationship between age and 10-year risk of hip fracture for men and women. This technique is designed to produce a smooth fit to the data that also reduces the influence of extreme outliers 32. Analyses were conducted stratifying by sex because it was expected that hip fracture risk, evaluation and management of osteoporosis would vary by sex.…”

Section: Methodsmentioning

confidence: 99%

Suboptimal osteoporosis evaluation and treatment in older men with and without additional high-risk factors for fractures

Narla

Hirano²,

et al. 2019

Journal of Investigative Medicine

View full text Add to dashboard Cite

We compared osteoporosis case-finding, evaluation and treatment in groups of Older Men and Older Women with age alone as a significant risk for fracture and Older Men with Higher Risk (older men additionally having previous hip fracture, corticosteroid use or androgen deprivation therapy). We studied 13,704 older men and women (≥70 years old) receiving care at a Veterans Affairs medical center from January 2000 to August 2010 whose 10-year hip fracture risk was assessed by limited FRAX score. The main outcome measures were the proportion of patients who had bone mineral density (by dual-energy X-ray absorptiometry [DXA]) and serum 25-hydroxy vitamin D (25-OH D) measurements performed, and calcium/vitamin D or bisphosphonates prescribed. The proportion of men with a 10-year hip fracture risk ≥3% with age alone as a risk was 48% and 88% in men aged 75–79 and ≥80 years, respectively. Compared with Older Women, fewer Older Men underwent DXA (12% vs 63%, respectively) and 25-OH D measurements (18% vs 39%), and fewer received calcium/vitamin D (20% vs 63%) and bisphosphonate (5% vs 44%) prescriptions. In Older Men with Higher Risk category, the proportion of men with 10-year hip fracture risk ≥3% ranged from 69% to 95%. Despite a higher risk and expectation that this group would have greater case detection and screening, few Older Men with Higher risk underwent DXA screening (27%–36%) and 25-OH D measurements (23%–28%), and received fewer calcium/vitamin D (40%–50%) and bisphosphonate (13%–24%) prescriptions. Considering the known morbidity and mortality, our findings underscore the need for improved evaluation and management of osteoporosis in older men at high risk for fracture.

show abstract

Section: Methodsmentioning

confidence: 99%

Suboptimal osteoporosis evaluation and treatment in older men with and without additional high-risk factors for fractures

Narla

Hirano²,

et al. 2019

Journal of Investigative Medicine

View full text Add to dashboard Cite

show abstract

“…Deviations from a straight line indicate departures from normality. If the left end of pattern is below the line, and the right end of pattern is above the line, it indicates long tails at both ends of the data distribution (Chambers et al 1983). As shown in Figure 11, the probability plot of modal frequency for Haenam deck in the middle part follows the line, but shows an s-shape deviation at the ends, which means the variance of data is higher than anticipated for a normal distribution.…”

Section: Statiscal Analysis Of Modal Propertiesmentioning

confidence: 99%

Statistical analysis of modal properties of a cable-stayed bridge through long-term structural health monitoring with wireless smart sensor networks

Asadollahi

2016

SPIE Proceedings

View full text Add to dashboard Cite

Understanding the dynamic behavior of complex structures such as long-span bridges requires dense deployment of sensors. Traditional wired sensor systems are generally expensive and time-consuming to install due to cabling. With wireless communication and on-board computation capabilities, wireless smart sensor networks have the advantages of being low cost, easy to deploy and maintain and therefore facilitate dense instrumentation for structural health monitoring. A long-term monitoring project was recently carried out for a cable-stayed bridge in South Korea with a dense array of 113 smart sensors, which feature the world's largest wireless smart sensor network for civil structural monitoring. This paper presents a comprehensive statistical analysis of the modal properties including natural frequencies, damping ratios and mode shapes of the monitored cable-stayed bridge. Data analyzed in this paper is composed of structural vibration signals monitored during a 12-month period under ambient excitations. The correlation between environmental temperature and the modal frequencies is also investigated. The results showed the long-term statistical structural behavior of the bridge, which serves as the basis for Bayesian statistical updating for the numerical model.

show abstract

“…However, we feel that the commonly used approaches, e.g., correlation coefficients, only provide one single value indicating the difference, which fails to answer how they differ from each other. In our approach, in addition to the traditional difference values, we use the Q-Q plot [35], to capture the details. Quantiles are points taken at regular intervals from the cumulative density function (CDF) of a random variable.…”

Section: Comparing Regions and Propertiesmentioning

confidence: 99%

VesselMap: A web interface to explore multivariate vascular data

Tao

Huang

Qiu

et al. 2016

Computers & Graphics

View full text Add to dashboard Cite

a b s t r a c tHemodynamics plays a key role in the pathological evolution of vascular diseases such as vascular stenosis, plaques and aneurysms. Augmented visualization of relevant hemodynamic and geometric data accelerates engineering designs of diagnostic and therapeutic methods for those vascular diseases. However, four dimensional (4D) hemodynamic data are intrinsically complex. Consequently, exploration of this information has not been streamlined, despite its importance. In this paper, we propose a web interface to explore multivariate hemodynamic data, thereby facilitating the visualization of spatial and temporal relationships among those hemodynamic parameters of interest. The main thrust of this web system is the proposed VesselMap methodology, a 2D representation of vessel structures that provides an overview of all query results without any visual occlusions. Furthermore, in this framework, quantilequantile plot was also used to compare spatial variations in parameter distributions, thereby greatly aiding in extracting local characteristics of hemodynamic parameters.Using "patient-specific" computational fluid dynamics (CFD) simulations, realistic flow velocity fields were obtained. Based on those CFD-simulated velocity fields, we performed an empirical evaluation to confirm the usefulness of our approach for biomedical engineering applications. We concluded that the proposed VesselMap methodology allowed users to rapidly interact with hemodynamic data of interest, enhancing efficacy of information retrieval.

show abstract

Graphical Methods for Data Analysis

Cited by 420 publications

References 0 publications

Suboptimal osteoporosis evaluation and treatment in older men with and without additional high-risk factors for fractures

Suboptimal osteoporosis evaluation and treatment in older men with and without additional high-risk factors for fractures

Statistical analysis of modal properties of a cable-stayed bridge through long-term structural health monitoring with wireless smart sensor networks

VesselMap: A web interface to explore multivariate vascular data

Contact Info

Product

Resources

About