Graphical presentation of distributions of risk in screening

2005 ◽  
Vol 12 (3) ◽  
pp. 155-160 ◽  
Author(s):  
J K Morris ◽  
N J Wald
Keyword(s):  
Relative Frequency ◽  
Gaussian Approximation ◽  
Gaussian Model ◽  
Likelihood Ratios ◽  
Frequency Distributions ◽  
Risk Estimates ◽  
Screening Performance ◽  
Log Likelihood ◽  
Standard Deviations ◽  
Screening Marker

Objective: The screening performance of tests involving multiple markers is usually presented visually as two Gaussian relative frequency distributions of risk, one curve relating to affected and the other to unaffected individuals. If the distribution of the underlying screening markers is approximately Gaussian, risk estimates based on the same markers will usually also be approximately Gaussian. However, this approximation sometimes fails. Here we examine the circumstances when this occurs. Setting: A theoretical statistical analysis. Methods: Hypothetical log Gaussian relative distributions of affected and unaffected individuals were generated for three antenatal screening markers for Down's syndrome. Log likelihood ratios were calculated for each marker value and plots of the relative frequency distributions were compared with plots of Gaussian distributions based on the means and standard deviations of these log likelihood ratios. Results: When the standard deviations of the distributions of a perfectly Gaussian screening marker are similar in affected and unaffected individuals, the distributions of risk estimates are also approximately Gaussian. If the standard deviations differ materially, incorrectly assuming that the distributions of the risk estimates are Gaussian creates a graphical anomaly in which the distributions of risk in affected and unaffected individuals plotted on a continuous risk scale intersect in two places. This is theoretically impossible. Plotting the risk distributions empirically reveals that all individuals have an estimated risk above a specified value. For individuals with more extreme marker values, the risk estimates reverse and increase instead of continuing to decrease. Conclusion: It is useful to check whether a Gaussian approximation for the distribution of risk estimates based on a screening marker is valid. If the value of the marker level at which risk reversal occurs lies within the set truncation limits, these may need to be reset, and a Gaussian model may be inappropriate to illustrate the risk distributions.

Calculation of roc curves in multidimensional likelihood ratio based screening with Down's syndrome as a special case

1998 ◽  
Vol 5 (2) ◽  
pp. 57-62 ◽  
Author(s):  
S O Larsen ◽  
M Christiansen ◽  
B Nørgaard-Pedersen
Keyword(s):  
Monte Carlo Simulation ◽  
Monte Carlo ◽  
Programming Language ◽  
A Priori ◽  
Roc Curves ◽  
Likelihood Ratios ◽  
Screening Performance ◽  
Log Likelihood ◽  
Special Knowledge ◽  
Special Case

Objectives The development of algorithms and computer programs for the analysis of screening performance in situations with multiple normally (Gaussian) distributed selection markers and a priori risks depending on a stratification of the population. Methods The S-PLUS programming language was used to construct programs producing distributions of log likelihood ratios based on the Monte Carlo simulation. These distributions were used to construct programs for the calculation of roc curves, including a possible stratification of the population. Results S-PLUS programs for the analysis of screening performance are listed and described. The programs can be used without any special knowledge of S-PLUS. An example of the use of the programs is given.

Monitoring Parkinson's Disease from phonation improvement by Log Likelihood Ratios

2015 ◽  
Author(s):  
Pedro Gomez-Vilda ◽  
Agustin Alvarez-Marquina ◽  
Victoria Rodellar-Biarge ◽  
Victor Nieto-Lluis ◽  
Rafael Martinez-Olalla ◽  
...  
Keyword(s):  
Parkinson’S Disease ◽  
Parkinson's Disease ◽  
Likelihood Ratios ◽  
Log Likelihood

Probabilistic Evaluation of SMS Messages as Forensic Evidence

2013 ◽  
pp. 138-149
Author(s):  
Shunichi Ishihara
Keyword(s):  
Likelihood Ratio ◽  
Low Cost ◽  
Scientific Evidence ◽  
Short Message ◽  
Likelihood Ratios ◽  
Forensic Evidence ◽  
Probabilistic Evaluation ◽  
Popular Medium ◽  
Log Likelihood ◽  
Message Service

This study is one of the first likelihood ratio-based forensic text comparison studies in forensic authorship analysis. The likelihood-ratio-based evaluation of scientific evidence has started being adopted in many disciplines of forensic evidence comparison sciences, such as DNA, handwriting, fingerprints, footwear, voice recording, etc., and it is largely accepted that this is the way to ensure the maximum accountability and transparency of the process. Due to its convenience and low cost, short message service (SMS) has been a very popular medium of communication for quite some time. Unfortunately, however, SMS messages are sometimes used for reprehensible purposes, e.g., communication between drug dealers and buyers, or in illicit acts such as extortion, fraud, scams, hoaxes, and false reports of terrorist threats. In this study, the author performs a likelihood-ratio-based forensic text comparison of SMS messages focusing on lexical features. The likelihood ratios (LRs) are calculated in Aitken and Lucy’s (2004) multivariate kernel density procedure, and are calibrated. The validity of the system is assessed based on the magnitude of the LRs using the log-likelihood-ratio cost (Cllr). The strength of the derived LRs is graphically presented in Tippett plots. The results of the current study are compared with those of previous studies.

scholarly journals Maternal Serum Invasive Trophoblast Antigen (Hyperglycosylated hCG) as a Screening Marker for Down Syndrome during the Second Trimester

2004 ◽  
Vol 50 (10) ◽  
pp. 1804-1808 ◽  
Author(s):  
Glenn E Palomaki ◽  
Louis M Neveux ◽  
George J Knight ◽  
James E Haddow ◽  
Raj Pandian
Keyword(s):  
United States ◽  
Down Syndrome ◽  
False Positive Rate ◽  
The United States ◽  
Maternal Serum ◽  
Second Trimester ◽  
Marker Panel ◽  
Screening Performance ◽  
Β Hcg ◽  
Screening Marker

Abstract Background: Approximately two million pregnancies in the United States are screened for Down syndrome annually by use of second-trimester maternal serum markers. At present, a combination of four markers can identify 75% of affected pregnancies when 5% of screened women are classified as candidates for amniocentesis. Although not currently included in screening panels, invasive trophoblast antigen (ITA) is a promising screening marker in serum or urine in both the second and first trimesters. This study aims at better defining the screening performance of serum ITA in the second trimester. Methods: In an earlier study, serum samples from an unbiased sampling of 45 Down syndrome (cases) and 238 unaffected (control) pregnancies between 14 and 20 weeks of gestation were collected from various centers in the United States. Samples were aliquoted and stored at −20 °C for 8 years. We measured ITA in these samples and determined the screening performance both univariately and in combination with other screening markers. Results: The median ITA in Down syndrome pregnancies was >3.00 multiples of the median, higher than that found for human chorionic gonadotropin (hCG) or free β-hCG. At a 5% false-positive rate, ITA univariately detected 38% and 40% of Down syndrome pregnancies, respectively, when assigned by date of last menstrual period or ultrasound date. Modeling yielded rates of 45% and 48%. ITA correlated strongly with hCG and free β-hCG. When substituted for either of these in a multiple marker panel, ITA performed comparably. Conclusions: This study indicates that serum ITA is an effective marker for Down syndrome. It is highly correlated with both hCG and free β-hCG and could replace either of them in a multiple marker panel.

scholarly journals Repeatable Battery for the Assessment of Neuropsychological Status (RBANS): Normative Data for Older Adults

2019 ◽  
Vol 34 (8) ◽  
pp. 1356-1366 ◽  
Author(s):  
Michelle Olaithe ◽  
Michael Weinborn ◽  
Talitha Lowndes ◽  
Amanda Ng ◽  
Erica Hodgson ◽  
...  
Keyword(s):  
Normative Data ◽  
Neuropsychological Test ◽  
Female Gender ◽  
Frequency Distributions ◽  
Neuropsychological Status ◽  
Three Samples ◽  
Regression Techniques ◽  
Repeatable Battery ◽  
Standard Deviations ◽  
And Gender

Abstract Objective Provide updated older adult (ages 60+) normative data for the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS), Form A, using regression techniques, and corrected for education, age, and gender. Method Participants (aged 60–93 years; N = 415) were recruited through the Healthy Ageing Research Program (HARP), University of Western Australia, and completed Form A of the RBANS as part of a wider neuropsychological test battery. Regression-based techniques were used to generate normative data rather than means-based methods. This methodology allows for the control of demographic variables using continuous data. To develop norms, the data were assessed for: (1) normality; (2) associations between each subtest score and age, education, and gender; (3) the effect of age, education, and gender on subtest scores; and (4) residual scores which were converted to percentile distributions. Results Differences were noted between the three samples, some of which were small and may not represent a clinically meaningful difference. Younger age, more years of education, and female gender were associated with better scores on most subtests. Frequency distributions, means, and standard deviations were produced using unstandardized residual scores to remove the effects of age, education, and gender. Conclusions These normative data expand upon past work by using regression-based techniques to generate norms, presenting percentiles, as well as means and standard deviations, correcting for the effect of gender, and providing a free-to-use Excel macro to calculate percentiles.

The composition of milk Part I. Variation in the solids not fat, fat and protein content of cow’s milk, and their relationship

1927 ◽  
Vol 17 (1) ◽  
pp. 62-71 ◽  
Author(s):  
H. T. Cranfield ◽  
D. G. Griffiths ◽  
E. R. Ling
Keyword(s):  
Protein Content ◽  
Fat Content ◽  
Cow’S Milk ◽  
Frequency Distributions ◽  
Cow's Milk ◽  
Standard Deviations ◽  
Correlation Tables

1. Percentages of fat, solids not fat and protein were determined in over 700 samples of mixed milk from 15 herds during 1925–26. In the case of fat content, nine herds produced one or more samples below 3 per cent., one herd recording 25 per cent, of samples below this limit. With regard to solids not fat, twelve herds produced milk containing less than 8·5 per cent, on one or more occasions, the highest percentage of deficient samples recorded being 40.2. Frequency distributions of fat, solids not fat and protein percentages in the samples analysed, are given, together with standard deviations, and mean percentages with probable errors for these three constituents.3. Correlation tables of fat with solids not fat, and protein with solids not fat have been prepared, and graphs illustrating the variations are given.

Sign in / Sign up

Export Citation Format

Share Document