Percentile Plots Research Articles

Characteristics of wideband acoustic immittance in patients with middle-ear dysfunction.

Wideband acoustic immittance (WAI) measurements are a relatively new class of aural acoustic tests that have shown promise as useful tools for evaluating middle-ear status. A growing body of work has described WAI for infants, children, and adults with normal middle-ear function, but a relatively limited number of studies have investigated the influence of specific middle-ear disorders on WAI. Although emerging WAI research data show promising results, additional measurements from ears of patients with a variety of middle-ear disorders are needed. The aims of the present study were to (1) obtain WAI data from patients with a variety of middle-ear conditions, (2) examine relationships between WAI data relative to standard audiometric tests, and (3) identify patterns or profiles in WAI data within and among patients with different middle-ear conditions. A descriptive study. Participants included 30 children and two adults with a variety of middle-ear disorders who were recruited from clinical audiology settings. Experimental ambient and tympanometric WAI data were gathered along with standard audiometric test data as part of clinical audiology appointments. Single-subject and small-group data sets for ears of patients with suspected otitis media with effusion (sOME), pressure equalization (PE) tubes, negative tympanometric peak pressure (TPP), tympanoplasty, and cerumen impaction were obtained. Qualitative analysis and descriptive statistics (means and percentiles) were used to characterize the data. Group mean absorbance for ears in the sOME group was reduced across the majority of frequencies; absorbance for ears with negative TPP was also reduced, although to a lesser degree. Absorbance patterns for ears with PE tubes show even greater differences because of the effects of the tube and open middle-ear cavity. WAI from ears with a tympanoplasty and cerumen impaction was suggestive of middle-ear dysfunction in the presence of normal single-frequency tympanometry. Percentile (5th, 50th, and 95th) plots for absorbance and other WAI quantities of admittance level, admittance phase, and equivalent ear-canal volume calculated for the sOME, negative TPP, and PE tube groups showed distinctive profiles that might be useful for discrimination of different middle-ear pathologies. Quantities from wideband tympanometry, including 250- and 1000-Hz single-frequency admittance tympanometric data were also derived. WAI measurements for some middle-ear conditions were consistent with patterns and profiles previously reported in the literature. WAI profiles for other middle-ear conditions, which had not been previously reported in the literature, were presented. Relationships between WAI data and results of standard audiometric tests were generally consistent with expectations. However, in some cases, WAI measurements were suggestive of middle-ear dysfunction despite normal 226-Hz admittance tympanometric findings.

The use of property-scale portable X-ray fluorescence data in gold exploration: advantages and limitations

This study examines field portable X-ray fluorescence (pXRF) data for unsieved B- and C-horizon soil samples from two contrasting Au exploration programs in northern Canada and compares them to laboratory results from sieved samples to determine if a minimalist approach to sample preparation, equipment calibration and analysis produces results that are adequate for exploration purposes. The collection of 14 651 residual soil samples on detailed grids at the Whiskey Au project in the Black Hills of the Yukon Territory, Canada during the 2011 field season allows a comparison of pXRF data from unsieved samples with inductively-coupled plasma mass spectrometry (ICP-MS) data from the <100-µm grain size fraction digested in aqua regia. XRF data were generated on two devices: an Innov-X Delta handheld and a Niton FXL desk-top unit. Data for selected elements from the pXRF units show trends in gridded percentile plots comparable to the ICP-MS data, although the continuity of these trends reflects the degree to which the pathfinder element concentrations exceed the lower orders of detection. Acceptable merged gridded images for As and Cu from the two devices were generated without data levelling, whereas the Pb, Mo and Ni data require leveling prior to gridding in order to provide a reasonable fit to the laboratory data. Data for elements that occur in concentrations close to the lower limit of detection by pXRF, such as Sb, may provide some useful information but must be used with caution. Data for Fe also show a poor correlation, possibly due to heterogeneity of Fe in the samples, as well as due to incomplete digestion of all Fe-bearing minerals in some samples using an aqua regia digestion. A total of 680 till samples collected from regional and detailed sampling grids on the Kiyuk Lake Au property in southern Nunavut, Canada during the 2012 field season were analysed by pXRF and also by ICP-MS following an aqua regia digestion. pXRF data were generated on an Innov-X Delta using unsieved samples for comparison with ICP-MS data from the <70-µm grain size fraction. The pathfinder element As shows good agreement with the laboratory data, whereas Ni, Mo and Sb data give poor matches due to proximity of the data to the lower orders of detection by pXRF. The use of pXRF data on soil samples with no sample preparation other than drying and no instrument calibration is a robust approach for Au exploration using particular elements, such As and Cu, provided the data are monitored for consistency. Even greater compatibility between pXRF and laboratory data would be expected where sample preparation and site-specific calibrations are implemented. Other pathfinder elements for Au, including W, Sb, Bi, Ag and Te, as well as Au itself, typically occur at levels in soil either close to or below the lower order of detection of the current generation of field-portable analytical devices. The use of pXRF in Au exploration allows for quick decision-making and provides near-real time sampling guidance in the field where ‘fit for purpose’ data for suitable pathfinder and lithologically-controlled elements can be obtained.

Wave energy resources along the Hawaiian Island chain

Hawaii's access to the ocean and remoteness from fuel supplies has sparked an interest in ocean waves as a potential resource to meet the increasing demand for sustainable energy. The wave resources include swells from distant storms and year-round seas generated by trade winds passing through the islands. This study produces 10 years of hindcast data from a system of mesoscale atmospheric and spectral wave models to quantify the wind and wave climate as well as nearshore wave energy resources in Hawaii. A global WAVEWATCH III (WW3) model forced by surface winds from the Final Global Tropospheric Analysis (FNL) reproduces the swell and seas from the far field and a nested Hawaii WW3 model with high-resolution winds from the Weather Research Forecast (WRF) model capture the local wave processes. The Simulating Waves Nearshore (SWAN) model nested inside Hawaii WW3 provides data in coastal waters, where wave energy converters are being considered for deployment. The computed wave heights show good agreement with data from satellites and buoys. Bi-monthly median and percentile plots show persistent trade winds throughout the year with strong seasonal variation of the wave climate. The nearshore data shows modulation of the wave energy along the coastline due to the undulating volcanic island bathymetry and demonstrates its importance in selecting suitable sites for wave energy converters.

Patterns and cycles in the Climate Forecast System Reanalysis wind and wave data

The Climate Forecast System Reanalysis and the corresponding WAVEWATCH III hindcast datasets allow climatic interpretation of winds as well as their impacts on waves. In this paper, we analyze the continuous 31years of global wind and wave data in terms of climate patterns and cycles. Quarterly averages and percentile plots of the wind speed and wave height illustrate the seasonal pattern and distributions of extreme events, while the annual and inter-annual variability demonstrates the wind and wave climate. The data is correlated with published indices of known atmospheric cycles. The datasets show good correspondence with the Arctic Oscillation, Antarctic Oscillation, El Nino Southern Oscillation, and the Madden-Julian Oscillation in both the wind and wave fields. The results compare well with published climate studies on regional scales and provide important linkage to the global wave climate characteristics.

Atlas of global wave energy from 10 years of reanalysis and hindcast data

Development of ocean wave energy requires general knowledge of the wave climate as well as reliable estimates of the available power. A WAVEWATCH III hindcast study with surface winds from the Final Global Tropospheric Analysis (FNL) provides 10 years of spectral wave data for a thorough examination of the wave resource around the world. The fully assimilated FNL data provides accurate wind input for more reliable predictions of the wave conditions than the operational forecasts. The computed wave parameters show good correlation with Jason-1 altimetry and buoy measurements. The monthly median plots describe the seasonal variation of the global wave climate and the percentile plots indicate the persistence and severity of the wave conditions. Separate estimates of the power from the wind waves and primary swell provide a precise description of the resource in support of design and planning of wave energy converters.

Quetiapine versus clomipramine in the augmentation of selective serotonin reuptake inhibitors for the treatment of obsessive-compulsive disorder: a randomized, open-label trial

After 12 weeks of selective serotonin reuptake inhibitor (SSRI) monotherapy with inadequate response, 10 patients received clomipramine and 11 received quetiapine as augmentation agents of the SSRI. The primary outcome measure was the difference between initial and final scores of the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), rated in a blinded fashion, and the score of clinical global improvement (CGI-I). Statistical analyses were performed using nonparametric tests to evaluate treatment efficacy and the difference between treatment groups. Percentile plots were constructed with YBOCS scores from the clomipramine and quetiapine groups. Considering response a >or=35% reduction in the initial Y-BOCS score plus a rating of 'much improved' or 'very much improved' on CGI-I, four of eleven quetiapine patients and one out of ten clomipramine patients were classified as responders. The mean final Y-BOCS score was significantly lower than baseline in the quetiapine augmentation group (P = 0.023), but not in the clomipramine augmentation group (P = 0.503). The difference between groups showed a trend towards significance only at week 4, the mean Y-BOCS score being lower for those receiving quetiapine (P = 0.052). A difference between groups was also observed at week 4 according to percentile plots. These results corroborate previous findings of quetiapine augmentation efficacy in obsessive-compulsive disorder (OCD). Clomipramine augmentation did not produce a significant reduction in Y-BOCS scores. Higher target maximum dosages might have yielded different results.

Geographic Box Plots

Traditional box plots, described by Tukey in 1977, can be biased representations of geographic or spatial variables, since they do not take into account the areas associated with the elements of geographic variables. To help solve this problem, we propose and evaluate a spatially (areally) weighted box plot, a "geographic box plot," that can be used to describe a wide variety of geographic variables. Much of our discussion of box plots also applies to "traditional" and "geographic" histograms and percentile plots. Traditional and geographic box plots of one hypothetical and two real-world geographic variables are constructed and compared to illustrate our points, as well as to illustrate the often marked differences between traditional and geographic box plots of the same geographic variables. The real-world geographic variables considered here are human population density (in the year 2000) within the contiguous 48 United States and five-year-average (January 2001-December 2005) air temperature over the Arctic land surface. Our traditional box plot of population density over represented the small, high-density states, whereas our corresponding geographic box plot correctly depicted the geographic distribution of population density. A geographic box plot correctly accounted for the reduction in spherical grid-node areas with latitude over the Arctic land surface, whereas a traditional box plot of the spherical grid-node values did not. Our hope is that geographers and other scientists will make refinements to and find many more applications for our geographic box plots.

The Box-Percentile Plot

A variant of the boxplot is proposed in which the sides contain the information of a percentile plot (which is equivalent to the empirical cumulative distribution function). Unlike boxplots, there is no question about how long to draw the whiskers, nor is there loss of information due to grouping. Side-by-side comparisons of distributions are especially effective. In spite of including more detail, the impact on statistically-untrained readers remains similar to that of traditional boxplots.

Tracking filter performance with particle counting

Particle counting can be a sensitive tool for monitoring filter performance and improving the quality of finished water.Minor changes (1–2 mg/L) in the primary coagulant dosage can significantly affect particle counts in filtered water. Jar tests were inconclusive in selecting a coagulant dosage for source water of low turbidity (<5 ntu). Measurements of turbidity in filtered water were also not definitively altered by small changes in coagulant dosage. On‐line particle counting was successfully used in a 0.5 × 106–m3/d (135‐mgd) conventional water treatment plant to distinguish the effects on water quality of small changes in coagulant dosage and filter flow. Measurable changes in particle counts in filtered water were detected when the aluminum sulfate dosage ranged from 5 to 10 mg/L at filter loading rates of 4–9.5 m/h (1.7–4.0 gpm/sq ft). Particle count percentile plots and statistics (10th, 50th, 90th, 95th, and 98th percentiles) were valuable indicators of the performance of both individual filters and the overall plant.

Quantification of the Relationship Between Insulin Sensitivity and β-Cell Function in Human Subjects: Evidence for a Hyperbolic Function

To determine the relationship between insulin sensitivity and beta-cell function, we quantified the insulin sensitivity index using the minimal model in 93 relatively young, apparently healthy human subjects of varying degrees of obesity (55 male, 38 female; 18-44 yr of age; body mass index 19.5-52.2 kg/m2) and with fasting glucose levels < 6.4 mM. SI was compared with measures of body adiposity and beta-cell function. Although lean individuals showed a wide range of SI, body mass index and SI were related in a curvilinear manner (P < 0.0001) so that on average, an increase in body mass index was associated generally with a lower value for SI. The relationship between the SI and the beta-cell measures was more clearly curvilinear and reciprocal for fasting insulin (P < 0.0001), first-phase insulin response (AIRglucose; P < 0.0001), glucose potentiation slope (n = 56; P < 0.005), and beta-cell secretory capacity (AIRmax; n = 43; P < 0.0001). The curvilinear relationship between SI and the beta-cell measures could not be distinguished from a hyperbola, i.e., SI x beta-cell function = constant. This hyperbolic relationship described the data significantly better than a linear function (P < 0.05). The nature of this relationship is consistent with a regulated feedback loop control system such that for any difference in SI, a proportionate reciprocal difference occurs in insulin levels and responses in subjects with similar carbohydrate tolerance. We conclude that in human subjects with normal glucose tolerance and varying degrees of obesity, beta-cell function varies quantitatively with differences in insulin sensitivity.(ABSTRACT TRUNCATED AT 250 WORDS)

Macroparasitic epizootic disease: a potential mechanism for the termination of sea urchin outbreaks in Northern Norway?

Outbreak populations of the green sea urchin, Strongylocentrotus droebachiensis Muller, in Vestfjorden, Northern Norway, are infected by the endoparasitic nematode, Echinomermella matsi Jones and Hagen, 1987. The prevalence of E. matsi has increased from 5.5% in 1983 to 65.4% in 1991 at a study site in Godoystraumen. Infected sea urchins had a lower mean gonad index than non-infected sea urchins, and two thirds of the infected sea urchins were reduced to virtual castrates of unknown sex. The mean density of the S. droebachiensis-population in Godoystraumen appeared unchanged, but the mean test diameter had decreased by 14.6%, from 36.4 mm in 1983 to 31.1 mm in 1991. As a result of these changes the reproductive capacity of the sea urchin population appears to have been reduced by approximately 58%. The data analysis was augmented by the use of a PEPLOM (PErcentile PLOt Matrix), which appears to be a novel combination of two powerful methods of graphical data analysis, i.e., the percentile comparison graph and the scatterplot matrix (SPLOM). Changes in the size-frequency distribution, and in the size-dependent pattern of parasite infection in the S. droebachiensis-population in Godoystraumen, are consistent with a hypothesis of increased parasite-related mortality. The larger hypothesis, that E. matsi may function as a macroparasitic terminator of sea urchin outbreaks in northern Norway, could not be rejected.

Probing depth, attachment loss and gingival recession. Findings from a clinical examination in Ushiku, Japan.

The present investigation describes probing pocket depth, probing attachment level and recession data from 319 randomly selected subjects, aged 20-79 years, from Ushiku, Japan. The findings are reported as mean values, frequency distributions and percentile plots of the 3 parameters at buccal, interproximal and lingual surfaces of single rooted (incisors, canines, premolars) and molar teeth. Inter-as well as intra-examiner errors for probing pocket depth and probing attachment levels were assessed and found to be small. The data reported revealed that practically all subjects studied had one or more sites in the dentition affected by destructive periodontal disease and that the severity of disease increased with age. It was further observed that in each age group, molars had suffered more attachment loss than single rooted teeth and that the interproximal surfaces as a rule had lost more periodontal tissue support than corresponding buccal and lingual surfaces. The attachment loss difference observed between different surfaces of a given tooth or a group of teeth, however, was comparatively small. In the age groups between 20-59 years, advanced destructive periodontal disease was found in a small subgroup of the subject sample, while after the age of 60 years, widespread destructive periodontitis was common. An attempt was made to examine the progression of destructive disease with age by comparing the frequency distributions of sites with attachment loss of greater than or equal to 3 mm in subjects of different age groups. The data suggested that in younger subject groups, progression was confined to a subset of individuals, while in older age groups, more subjects and sites became involved. A major feature of destructive periodontal disease in older individuals was the accompaniment of attachment loss with recession at the gingival margin. Deep pockets were relatively infrequently detected, while advanced loss of attachment (with recession) occurred at many sites.

Methods of evaluating periodontal disease data in epidemiological research.

The purpose of the present investigation was to examine the use of different methods of data analysis to examine the pattern of periodontal tissue destruction in a group of adults. A clinical examination was carried out on 319 subjects, 20-79 years of age, randomly selected from the population of the city of Ushiku, Japan. The following parameters were studied: missing teeth, plaque, gingivitis, probing pocket depth and probing attachment level. All surfaces of all teeth were examined. The data obtained were evaluated in 3 ways: (i) mean values for each of the clinical parameters were computed for each subject and age group, (ii) the frequency of subjects with one or more sites of attachment loss above certain thresholds was assessed, (iii) percentile plots were generated in such a way that for each age group the percent of sites in subjects with varying attachment levels could be determined. The results revealed that the subjects included in the study had poor oral hygiene, high frequency of gingivitis and loss of probing attachment which increased with age. The mean annual attachment loss for all tooth surfaces was calculated to be less than 0.1 mm. Severe periodontal tissue breakdown was not as common as might have been expected. Hence, the majority of the subjects in the 20-59 year age group had very little evidence of destructive periodontal disease. In these age groups, a comparatively small subfraction accounted for most of the disease detected. After the age of 60, periodontal disease appeared to be more widespread among the subjects examined.(ABSTRACT TRUNCATED AT 250 WORDS)

Estimating reference ranges in clinical pathology: an objective approach.

Conventional reference ranges evolve from subjective criteria for health and disease. We offer an objective method for distinguishing typical from atypical values by purely statistical criteria. We define typical values as those exhibiting a linear relationship with percentiles on a value versus percentile plot. Identification of percentiles at which deviation from linearity occurs results from calculation of correlation coefficients between values and percentiles over centrally expanding ranges of percentiles. One selects arbitrarily some minimum value for these correlation coefficients, for example, 0.990, as the criterion for deviation from linearity. Values encompassed by these percentiles of deviation constitute an objective reference range. Identification of any correlations between atypical values and symptoms of disease requires clinical follow-up studies.

A Method of Correcting for Linear Drift in Computed Tomography Brain Scans

Linear drift of X-ray attenuation coefficients must be corrected if quantitative comparisons are to be made between computed tomography (CT) brain scans of the same individual performed at different times. Such a correction is accomplished by comparing the low (cerebrospinal fluid) end of the attenuation coefficient frequency histograms using a percentile--percentile plot. A "drift correction" permits serial quantitative assessments of the progression or regression of white matter hypodensity, such as occurs in drug induced leukoencephalopathy.

Phonemic errors of deaf children on a nonsense syllable test

A closed response nonsense syllable test was developed to assess the reception of gross phonemic contrasts by deaf children. Test items were composed of V, VC, and CV syllables with the vowels /i,a,u/ and the consonants /m,s,g/. The test was administered over a two‐year period (1975 and 1976) to one hundred 12–14 year old deaf children enrolled in schools for the deaf throughout New York State. Percentile plots showed improvement in overall performance over the two‐year period. Greatest improvements were for vowels. Results indicated a pattern of errors similar to those reported by previous investigators for adults and adolescents [Pickett et al., in Speech Communication Ability and Profound Deafness, 119–133 (1970)]. These included: (1) scores for vowels were consistently higher than those for consonants, (2) scores for /u/ and /a/ were roughly the same and were in turn higher than the scores for /i/, (3) scores for the initial consonants were higher than scores for final consonants except for the nasal in which the reverse was true. [Work supported by Cooperative Research Endeavors in Education of the Deaf (CREED), Bureau for Special School Supervision, Office for Education of Children with Handicapping Conditions].

Accuracy Specifications for Automatic Omega Navigators

Fully automatic Omega navigation systems have been demonstrated and are beginning to be used operationally in aircraft and other vehicles. A natural question is the accuracy achieved by such systems. Concern over the probability of maintaining correct lane identification complicates this question. Use of the root-mean-square (rms) statistics is shown to be a misleading measure of operational performance. Alternate methods of specifying the navigational errors of automatic Omega navigators are discussed. Cumulative percentile plots are recommended for the specification of errors. Use of Weibull probability paper to study radial error trends is demonstrated. Illustrations of error analysis techniques are given, based on actual Omega flight test data.