A Software Tool for Estimating Uncertainty of Bayesian Posterior Probability for Disease

The role of medical diagnosis is essential in patient care and healthcare. Established diagnostic practices typically rely on predetermined clinical criteria and numerical thresholds. In contrast, Bayesian inference provides an advanced framework that supports diagnosis via in-depth probabilistic analysis. This study's aim is to introduce a software tool dedicated to the quantification of uncertainty in Bayesian diagnosis, a field that has seen minimal exploration to date. The presented tool, a freely available specialized software program, utilizes uncertainty propagation techniques to estimate the sampling, measurement, and combined uncertainty of the posterior probability for disease. It features two primary modules and fifteen submodules, all designed to facilitate the estimation and graphical representation of the standard uncertainty of the posterior probability estimates for diseased and non-diseased population samples, incorporating parameters such as the mean and standard deviation of the test measurand, the size of the samples, and the standard measurement uncertainty inherent in screening and diagnostic tests. Our study showcases the practical application of the program by examining the fasting plasma glucose data sourced from the National Health and Nutrition Examination Survey. Parametric distribution models are explored to assess the uncertainty of Bayesian posterior probability for diabetes mellitus, using the oral glucose tolerance test as the reference diagnostic method.

Embracing uncertainty: The way forward in plant fossil phylogenetics

Although molecular phylogenetics remains the most widely used method of inferring the evolutionary history of living groups, the last decade has seen a renewed interest in morphological phylogenetics, mostly driven by the promises that integrating the fossil record in phylogenetic trees offers to our understanding of macroevolutionary processes and dynamics and the possibility that the inclusion of fossil taxa could lead to more accurate phylogenetic hypotheses. The plant fossil record presents some challenges to its integration in a phylogenetic framework. Phylogenies including plant fossils often retrieve uncertain relationships with low support, or lack of resolution. This low support is due to the pervasiveness of morphological convergence among plant organs and the fragmentary nature of many plant fossils, and it is often perceived as a fundamental weakness reducing the utility of plant fossils in phylogenetics. Here I discuss the importance of uncertainty in morphological phylogenetics and how we can identify important information from different patterns and types of uncertainty. I also review a set of methodologies that can allow us to understand the causes underpinning uncertainty and how these practices can help us to further our knowledge of plant fossils. I also propose that a new visual language, including the use of networks instead of trees, represents an improvement on the old visualization based on consensus trees and more adequately serves phylogeneticists working with plant fossils. This set of methods and visualization tools represents an important way forward in a fundamental field for our understanding of the evolutionary history of plants.

A Bayesian Inference Based Computational Tool for Parametric and Nonparametric Medical Diagnosis

Medical diagnosis is the basis for treatment and management decisions in healthcare. Conventional methods for medical diagnosis commonly use established clinical criteria and fixed numerical thresholds. The limitations of such an approach may result in a failure to capture the intricate relations between diagnostic tests and the varying prevalence of diseases. To explore this further, we have developed a freely available specialized computational tool that employs Bayesian inference to calculate the posterior probability of disease diagnosis. This novel software comprises of three distinct modules, each designed to allow users to define and compare parametric and nonparametric distributions effectively. The tool is equipped to analyze datasets generated from two separate diagnostic tests, each performed on both diseased and nondiseased populations. We demonstrate the utility of this software by analyzing fasting plasma glucose, and glycated hemoglobin A1c data from the National Health and Nutrition Examination Survey. Our results are validated using the oral glucose tolerance test as a reference standard, and we explore both parametric and nonparametric distribution models for the Bayesian diagnosis of diabetes mellitus.

Calibration of computational tools for missense variant pathogenicity classification and ClinGen recommendations for PP3/BP4 criteria

Recommendations from the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) for interpreting sequence variants specify the use of computational predictors as "supporting" level of evidence for pathogenicity or benignity using criteria PP3 and BP4, respectively. However, score intervals defined by tool developers, and ACMG/AMP recommendations that require the consensus of multiple predictors, lack quantitative support. Previously, we described a probabilistic framework that quantified the strengths of evidence (supporting, moderate, strong, very strong) within ACMG/AMP recommendations. We have extended this framework to computational predictors and introduce a new standard that converts a tool's scores to PP3 and BP4 evidence strengths. Our approach is based on estimating the local positive predictive value and can calibrate any computational tool or other continuous-scale evidence on any variant type. We estimate thresholds (score intervals) corresponding to each strength of evidence for pathogenicity and benignity for thirteen missense variant interpretation tools, using carefully assembled independent data sets. Most tools achieved supporting evidence level for both pathogenic and benign classification using newly established thresholds. Multiple tools reached score thresholds justifying moderate and several reached strong evidence levels. One tool reached very strong evidence level for benign classification on some variants. Based on these findings, we provide recommendations for evidence-based revisions of the PP3 and BP4 ACMG/AMP criteria using individual tools and future assessment of computational methods for clinical interpretation.

Bayesian two-stage design for phase II oncology trials with binary endpoint

In phase II oncology trials, two-stage design allowing early stopping for futility and/or efficacy is frequently used. However, this design based on frequentist statistical approaches could not guarantee a high posterior probability of attending the pre-specified clinically interesting rate from a Bayesian perspective. Here, we proposed a new Bayesian design enabling early terminating for efficacy as well as futility. In addition to the clinically uninteresting and interesting response rate, a prior distribution of response rate, the minimum posterior threshold probabilities and the lengths of the highest posterior density intervals were specified in the design. Finally, we defined the feasible design with the highest total effective predictive probability. We studied the properties of the proposed design and applied it to an oncology trial as an example. The proposed design ensured that the observed response rate fell within prespecified levels of posterior probability. The proposed design provides an alternative design to single-arm two-stage trials.

How Accurately Does the Information on Motor Development Collected During Health Checkups for Infants Predict the Diagnosis of Neurodevelopmental Disorders? - A Bayesian Network Model-Based Study

PURPOSE

We investigated to what extent early motor development problems predict a future diagnosis of neurodevelopmental disorders (NDDs)/Early Symptomatic Syndromes Eliciting Neurodevelopmental Examinations (ESSENCE) by using a Bayesian network model (BN).

SUBJECTS AND METHODS

Subjects were the children who had participated in the 18- and 36-month checkups in two cities in Japan between April 2014 and March 2015. Their motor development data at the 4-, 10- and 18-month-checkups were collected with ethical consideration. The diagnosis was confirmed at the age of six, after regular assessment in all developmental areas at a neurodevelopmental clinic. The accuracy of prediction of NDD based on posterior probabilities determined using the BN was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC). The posterior probability (the optimal cut-off value) yielding the maximum Youden Index (sensitivity + specificity - 1) is determined with the ROC curve, and the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and utility index (UI) were computed.

RESULTS

BN models showed associations between early motor items and developmental coordination disorders, borderline intelligence/intellectual disability, and speech and language disorder. The ROC curve for any NDD had an AUC of 0.735. The posterior probability with the maximal Youden Index was 0.138; at the optimal cut-off value, the sensitivity, specificity, PPV, NPV, UI+, and UI- were 0.619, 0.761, 0.250, 0.940, 0.155 and 0.715, respectively.

CONCLUSION

We utilized a novel approach in detailing the associations between certain early motor problems and specific NDDs. We showed that the presence of motor development problems early in development increases the probability of a future diagnosis of any NDD. Still, the sensitivity of early motor development problems as a screening tool was not high enough to be the sole instrument for detecting NDDs. The need for a broad, holistic ESSENCE perspective when looking at the course of motor development problems was stressed.

Applying posterior probability informed thresholds to traditional cranial trait sex estimation methods

Sex estimation methods using traditional cranial nonmetric traits utilize predictive models to produce a final sex estimation, using the resulting model's score to classify the individual. When sex estimations are assigned from discriminant scoring alone, statistical confidence in the resultant estimate is not always assessed or reported. Although some forensic anthropologists may qualitatively report their confidence in the assessment (e.g., "probable male"), these statements are subjective, not standardized, and not necessarily based on statistical results in a uniform way. The goals of this study were to evaluate how posterior probability-informed thresholds (PPITs) impacted accuracy rates, assess the balance between sample inclusion and accuracy for the proposed PPIT approach, and make recommendations for the use and interpretation of specific thresholds in casework. Using a sample of U.S. Black and White females and males (n = 292), we examined how PPITs can standardize the decision-making process of inferring sex for two methods using nonmetric cranial traits. We found that using PPITs of at least 0.85 increased accuracy (over 92% for some PPITs) yet remained highly inclusive of the sample. PPITs < 0.75 did not produce classification accuracy rates significantly higher than chance, and when using these cranial trait sex estimation methods, cases with posterior probabilities (PPs) <0.75 should be reported as "indeterminate." The 0.75-0.84 PPIT interval had an accuracy rate of 76%, which was both statistically significantly different from chance as well as from the higher (>0.85) groups, suggesting that although sex estimation at this level may be acceptable, the results hold lower confidence.

Bayesian Statistics for Surgical Decision Making

Background: Application of clinical study findings to surgical decision making requires accurate interpretation of the results, integration of the findings within the context of pre-existing knowledge and use of statistics to answer clinically relevant questions. Bayesian analyses are optimally suited for interpretation of study findings, supporting translation to the bedside. Discussion: Surgical decision making is a complex process that draws on an individual clinician's medical knowledge, experience, data, and the patient's unique characteristics and preferences. Subjective and objective knowledge may be merged to derive a probability of benefit or harm of a treatment under consideration. Bayesian reasoning complements the clinical decision-making process by incorporating known evidence and data from a new study to determine the probability of an outcome of interest. Bayesian analyses are statistically robust and intuitive when translating findings of a study into clinical care. In contrast, frequentist statistics are poorly suited to translate study findings to clinical application. This review aims to highlight the benefits of incorporating Bayesian analyses into clinical research. Conclusion: Bayesian analyses offer clinically relevant information including the probability of benefit or harm of a treatment under consideration while accounting for uncertainty. This information may be incorporated easily and accurately into surgical decision making.

Clinical Value of SARS-CoV2 IgM and IgG Antibodies in Diagnosis of COVID-19 in Suspected Cases

OBJECTIVE

To explore the clinical value of SARS-CoV2 IgM and IgG antibodies in the diagnosis of COVID-19 in suspected cases by likelihood ratio.

METHODS

By reinterpreting data from a previous study, the positive likelihood ratio of IgM and IgG antibodies in COVID-19 pneumonia diagnosis was calculated, and the posterior probability of IgM and IgG antibodies and their tandem detection was calculated finally.

RESULTS

The positive likelihood ratios of single IgM and IgG antibodies were 18.50 and 12.65, respectively, and the posterior probabilities were 90.18% and 86.26%, respectively. However, the posterior probability of the two antibody-tandem test was 99.15%, which could give clinicians more quantitative confidence in the diagnosis of COVID-19 in suspected cases. According to the results of this study, combining the advantages and disadvantages of nucleic acid testing and antibody detection, a feasible clinical path was found for clinicians to diagnose COVID-19 pneumonia from suspected cases.

CONCLUSION

For suspected cases, IgM- and IgG-antibody tests should first be done at the same time. If all antibody tests are positive, COVID-19 pneumonia could be confirmed. If not, nucleic acid detection (once or more) should be carried out, and in extreme cases high-throughput viral genome sequencing is required.

Identification of Pre-Clinical Alzheimer's Disease in a Population of Elderly Cognitively Normal Participants

Alzheimer's disease (AD) has a long pathological process, with an approximate lead-time of 20 years. During the early stages of the disease process, little evidence of the building pathology is identifiable without cerebrospinal fluid and/or imaging analyses. Clinical manifestations of AD do not present until irreversible pathological changes have occurred. Given an opportunity to provide treatment prior to irreversible pathological change, this study aims to identify a subgroup of cognitively normal (CN) participants from the Australian Imaging, Biomarker & Lifestyle Flagship Study of Ageing (AIBL), where subtle changes in cognition are indicative of early AD-related pathology. Using a Bayesian method for unsupervised clustering via mixture models, we define an aggregate measure of posterior probabilities (AMPP score) establishing the likelihood of pre-clinical AD. From Baseline through to 54 months, visuo-spatial function had the greatest contribution to the AMPP score, followed by attention and processing speed and visual memory. Participants with the highest AMPP scores had both increasing neo-cortical amyloid burden and decreasing hippocampus volume over 54 months, compared to those in the lowest category with stable amyloid burden and hippocampus volume. The identification of a possible pre-clinical stage in CN participants via this method, without the aid of disease specific biomarkers, represents an important step in utilizing the strength of cognitive composite scores for the early detection of AD pathology.

Group Testing-Based Robust Algorithm for Diagnosis of COVID-19

At the time of writing, the COVID-19 infection is spreading rapidly. Currently, there is no vaccine or treatment, and researchers around the world are attempting to fight the infection. In this paper, we consider a diagnosis method for COVID-19, which is characterized by a very rapid rate of infection and is widespread. A possible method for avoiding severe infections is to stop the spread of the infection in advance by the prompt and accurate diagnosis of COVID-19. To this end, we exploit a group testing (GT) scheme, which is used to find a small set of confirmed cases out of a large population. For the accurate detection of false positives and negatives, we propose a robust algorithm (RA) based on the maximum a posteriori probability (MAP). The key idea of the proposed RA is to exploit iterative detection to propagate beliefs to neighbor nodes by exchanging marginal probabilities between input and output nodes. As a result, we show that our proposed RA provides the benefit of being robust against noise in the GT schemes. In addition, we demonstrate the performance of our proposal with a number of tests and successfully find a set of infected samples in both noiseless and noisy GT schemes with different COVID-19 incidence rates.

Wireless Body Area Network (WBAN)-Based Telemedicine for Emergency Care

This paper is a collection of telemedicine techniques used by wireless body area networks (WBANs) for emergency conditions. Furthermore, Bayes’ theorem is proposed for predicting emergency conditions. With prior knowledge, the posterior probability can be found along with the observed evidence. The probability of sending emergency messages can be determined using Bayes’ theorem with the likelihood evidence. It can be viewed as medical decision-making, since diagnosis conditions such as emergency monitoring, delay-sensitive monitoring, and general monitoring are analyzed with its network characteristics, including data rate, cost, packet loss rate, latency, and jitter. This paper explains the network model with 16 variables, with one describing immediate consultation, as well as another three describing emergency monitoring, delay-sensitive monitoring, and general monitoring. The remaining 12 variables are observations related to latency, cost, packet loss rate, data rate, and jitter.

An Impact Location Algorithm for Spacecraft Stiffened Structure Based on Posterior Possibility Correlation

In order to ensure the safety of spacecrafts in orbit, impact location is an important part of structural health monitoring systems. In this paper, an impact location algorithm based on posterior probability correlation is proposed to solve the problem, that is, the impact point in the stiffened structure of a spacecraft is difficult to locate. The algorithm combines the Gaussian cross-correlation possibility weight method and the Bayesian posterior probability method. The cross-correlation possibility weight superposition based on grids was used to reduce the dependence of the accuracy of time difference extraction. Gaussian and normalized fitting were used to compensate the reflection, modal transformation, and amplitude attenuation of a stiffened plate. The location result was further optimized by the posterior probability. The proposed algorithm can be applied to the impact source localization of complex stiffened plate structures. The experiment results showed that the average location error can be 2.57 cm with proper sensor network schemes.

Bayesian sample-size determination methods considering both worthwhileness and unpromisingness for exploratory two-arm randomized clinical trials with binary endpoints

A randomized exploratory clinical trial comparing an experimental treatment with a control treatment on a binary endpoint is often conducted to make a go or no-go decision. Such an exploratory trial needs to have an adequate sample size such that it will provide convincing evidence that the experimental treatment is either worthwhile or unpromising relative to the control treatment. In this paper, we propose three new sample-size determination methods for an exploratory trial, which utilize the posterior probabilities calculated from predefined efficacy and inefficacy criteria leading to a declaration of the worthwhileness or unpromisingness of the experimental treatment. Simulation studies, including numerical investigation, showed that all three methods could declare the experimental treatment as worthwhile or unpromising with a high probability when the true response probability of the experimental treatment group is higher or lower, respectively, than that of the control treatment group.

Statistical parametric mapping of biomechanical one-dimensional data with Bayesian inference

Recent developments in Statistical Parametric Mapping (SPM) for continuum data (e.g. kinematic time series) have been adopted by the biomechanics research community with great interest. The Python/MATLAB package spm1d developed by T. Pataky has introduced SPM into the biomechanical literature, adapted originally from neuroimaging. The package already allows many of the statistical analyses common in biomechanics from a frequentist perspective. In this paper, we propose an application of Bayesian analogs of SPM based on Bayes factors and posterior probability with default priors using the BayesFactor package in R. Results are provided for two typical designs (two-sample and paired sample t-tests) and compared to classical SPM results, but more complex standard designs are possible in both classical and Bayesian frameworks. The advantages of Bayesian analyses in general and specifically for SPM are discussed. Scripts of the analyses are available as supplementary materials.

Bayesian selection of misspecified models is overconfident and may cause spurious posterior probabilities for phylogenetic trees

The Bayesian method is noted to produce spuriously high posterior probabilities for phylogenetic trees in analysis of large datasets, but the precise reasons for this overconfidence are unknown. In general, the performance of Bayesian selection of misspecified models is poorly understood, even though this is of great scientific interest since models are never true in real data analysis. Here we characterize the asymptotic behavior of Bayesian model selection and show that when the competing models are equally wrong, Bayesian model selection exhibits surprising and polarized behaviors in large datasets, supporting one model with full force while rejecting the others. If one model is slightly less wrong than the other, the less wrong model will eventually win when the amount of data increases, but the method may become overconfident before it becomes reliable. We suggest that this extreme behavior may be a major factor for the spuriously high posterior probabilities for evolutionary trees. The philosophical implications of our results to the application of Bayesian model selection to evaluate opposing scientific hypotheses are yet to be explored, as are the behaviors of non-Bayesian methods in similar situations.

Bayes Factors Unmask Highly Variable Information Content, Bias, and Extreme Influence in Phylogenomic Analyses

As the application of genomic data in phylogenetics has become routine, a number of cases have arisen where alternative data sets strongly support conflicting conclusions. This sensitivity to analytical decisions has prevented firm resolution of some of the most recalcitrant nodes in the tree of life. To better understand the causes and nature of this sensitivity, we analyzed several phylogenomic data sets using an alternative measure of topological support (the Bayes factor) that both demonstrates and averts several limitations of more frequently employed support measures (such as Markov chain Monte Carlo estimates of posterior probabilities). Bayes factors reveal important, previously hidden, differences across six "phylogenomic" data sets collected to resolve the phylogenetic placement of turtles within Amniota. These data sets vary substantially in their support for well-established amniote relationships, particularly in the proportion of genes that contain extreme amounts of information as well as the proportion that strongly reject these uncontroversial relationships. All six data sets contain little information to resolve the phylogenetic placement of turtles relative to other amniotes. Bayes factors also reveal that a very small number of extremely influential genes (less than 1% of genes in a data set) can fundamentally change significant phylogenetic conclusions. In one example, these genes are shown to contain previously unrecognized paralogs. This study demonstrates both that the resolution of difficult phylogenomic problems remains sensitive to seemingly minor analysis details and that Bayes factors are a valuable tool for identifying and solving these challenges.

Typicality and Predictive Distributions in Discriminant Function Analysis

While discriminant function analysis is an inherently Bayesian method, researchers attempting to estimate ancestry in human skeletal samples often follow discriminant function analysis with the calculation of frequentist-based typicalities for assigning group membership. Such an approach is problematic because it fails to account for admixture and for variation in why individuals may be classified as outliers or nonmembers of particular groups. This article presents an argument and methodology for employing a fully Bayesian approach in discriminant function analysis applied to cases of ancestry estimation. The approach requires adding the calculation, or estimation, of predictive distributions as the final step in ancestry-focused discriminant analyses. The methods for a fully Bayesian multivariate discriminant analysis are illustrated using craniometrics from identified population samples within the Howells published data. The article also presents ways to visualize predictive distributions calculated in more than three dimensions, explains the limitations of typicality measures, and suggests an analytical route for future studies of ancestry and admixture based in discriminant function analysis.

Using previous trial results to inform hypothesis testing of new interventions

Results of industry-sponsored Phase III trials registered at clinicaltrials.gov include a rich amount of information on the efficacy of medical interventions. We propose that these results can be used to inform hypothesis testing of a new intervention through the Bayes principle. The posterior probability of positive efficacy offers an accessible interpretation of the uncertainty of efficacy and a convenient metric for global false-positive control.

Bayesian noninferiority test for 2 binomial probabilities as the extension of Fisher exact test

Noninferiority trials have recently gained importance for the clinical trials of drugs and medical devices. In these trials, most statistical methods have been used from a frequentist perspective, and historical data have been used only for the specification of the noninferiority margin Δ>0. In contrast, Bayesian methods, which have been studied recently are advantageous in that they can use historical data to specify prior distributions and are expected to enable more efficient decision making than frequentist methods by borrowing information from historical trials. In the case of noninferiority trials for response probabilities π₁ ,π₂ , Bayesian methods evaluate the posterior probability of H₁ :π₁ >π₂ -Δ being true. To numerically calculate such posterior probability, complicated Appell hypergeometric function or approximation methods are used. Further, the theoretical relationship between Bayesian and frequentist methods is unclear. In this work, we give the exact expression of the posterior probability of the noninferiority under some mild conditions and propose the Bayesian noninferiority test framework which can flexibly incorporate historical data by using the conditional power prior. Further, we show the relationship between Bayesian posterior probability and the P value of the Fisher exact test. From this relationship, our method can be interpreted as the Bayesian noninferior extension of the Fisher exact test, and we can treat superiority and noninferiority in the same framework. Our method is illustrated through Monte Carlo simulations to evaluate the operating characteristics, the application to the real HIV clinical trial data, and the sample size calculation using historical data.

Discovery of Causal Models that Contain Latent Variables through Bayesian Scoring of Independence Constraints

Discovering causal structure from observational data in the presence of latent variables remains an active research area. Constraint-based causal discovery algorithms are relatively efficient at discovering such causal models from data using independence tests. Typically, however, they derive and output only one such model. In contrast, Bayesian methods can generate and probabilistically score multiple models, outputting the most probable one; however, they are often computationally infeasible to apply when modeling latent variables. We introduce a hybrid method that derives a Bayesian probability that the set of independence tests associated with a given causal model are jointly correct. Using this constraint-based scoring method, we are able to score multiple causal models, which possibly contain latent variables, and output the most probable one. The structure-discovery performance of the proposed method is compared to an existing constraint-based method (RFCI) using data generated from several previously published Bayesian networks. The structural Hamming distances of the output models improved when using the proposed method compared to RFCI, especially for small sample sizes.

Estimating the Probability of Traditional Copying, Conditional on Answer-Copying Statistics

Statistics for detecting copying on multiple-choice tests produce p values measuring the probability of a value at least as large as that observed, under the null hypothesis of no copying. The posterior probability of copying is arguably more relevant than the p value, but cannot be derived from Bayes' theorem unless the population probability of copying and probability distribution of the answer-copying statistic under copying are known. In this article, the authors develop an estimator for the posterior probability of copying that is based on estimable quantities and can be used with any answer-copying statistic. The performance of the estimator is evaluated via simulation, and the authors demonstrate how to apply the formula using actual data. Potential uses, generalizability to other types of cheating, and limitations of the approach are discussed.

Fast Coalescent-Based Computation of Local Branch Support from Quartet Frequencies

Species tree reconstruction is complicated by effects of incomplete lineage sorting, commonly modeled by the multi-species coalescent model (MSC). While there has been substantial progress in developing methods that estimate a species tree given a collection of gene trees, less attention has been paid to fast and accurate methods of quantifying support. In this article, we propose a fast algorithm to compute quartet-based support for each branch of a given species tree with regard to a given set of gene trees. We then show how the quartet support can be used in the context of the MSC to compute (1) the local posterior probability (PP) that the branch is in the species tree and (2) the length of the branch in coalescent units. We evaluate the precision and recall of the local PP on a wide set of simulated and biological datasets, and show that it has very high precision and improved recall compared with multi-locus bootstrapping. The estimated branch lengths are highly accurate when gene tree estimation error is low, but are underestimated when gene tree estimation error increases. Computation of both the branch length and local PP is implemented as new features in ASTRAL.

Diagnosis of Alzheimer's disease using neuropsychological testing improved by multivariate analyses

Neuropsychological assessment aids in the diagnosis of Alzheimer's disease (AD) by objectively establishing cognitive impairment from standardized tests. We present new criteria for diagnosis that use weighted combined scores from multiple tests. Our method employs two multivariate analyses: principal components analysis (PCA) and discriminant analysis. PCA (N = 216 participants) created more interpretable cognitive dimensions by resolving 49 test measures in our neuropsychological battery to 13 component scores for each participant. The component scores were used to build discriminant functions that classified each participant as either an early-stage AD (N = 55) or normal elderly (N = 78). Our discriminant function performed with high accuracy, sensitivity, and specificity (nearly all >90%) in the development, a cross-validation, and a new-subjects validation. When contrasted to two different traditional empirical methods for diagnosis (using cutscores and defining AD as falling below 5% on two or more test domains), our results suggested that the multivariate method was superior in classification (approximately 20% more accurate).