Narrow Results

Ensembles with several classifiers (such as neural networks or decision trees) are widely used to improve the generalization performance over a single classifier. Proper diversity among component classifiers is considered an important parameter for ensemble construction so that failure of one may be compensated by others. Among various approaches, data sampling, i.e., different data sets for different classifiers, is found more effective than other approaches. A number of ensemble methods have been proposed under the umbrella of data sampling in which some are constrained to neural networks or decision trees and others are commonly applicable to both types of classifiers. We studied prominent data sampling techniques for neural network ensembles, and then experimentally evaluated their effectiveness on a common test ground. Based on overlap and uncover, the relation between generalization and diversity is presented. Eight ensemble methods were tested on 30 benchmark classification problems. We found that bagging and boosting, the pioneer ensemble methods, are still better than most of the other proposed methods. However, negative correlation learning that implicitly encourages different networks to different training spaces is shown as better or at least comparable to bagging and boosting that explicitly create different training spaces.

An ensemble performs well when the component classifiers are diverse yet accurate, so that the failure of one is compensated for by others. A number of methods have been investigated for constructing ensemble in which some of them train classifiers with the generated patterns. This study investigates a new technique of training pattern generation. The method alters input feature values of some patterns using the values of other patterns to generate different patterns for different classifiers. The effectiveness of neural network ensemble based on the proposed technique was evaluated using a suite of 25 benchmark classification problems, and was found to achieve performance better than or competitive with related conventional methods. Experimental investigation of different input values alteration techniques finds that alteration with pattern values in the same class is better for generalization, although other alteration techniques may offer more diversity.

Polygenic risk scores (PRS) have led to enthusiasm for precision medicine. However, it is well documented that PRS do not generalize across groups differing in ancestry or sample characteristics e.g., age. Quantifying performance of PRS across different groups of study participants, using genome-wide association study (GWAS) summary statistics from multiple ancestry groups and sample sizes, and using different linkage disequilibrium (LD) reference panels may clarify which factors are limiting PRS transferability. To evaluate these factors in the PRS generation process, we generated body mass index (BMI) PRS (PRS_BMI) in the Electronic Medical Records and Genomics (eMERGE) network (N=75,661). Analyses were conducted in two ancestry groups (European and African) and three age ranges (adult, teenagers, and children). For PRS_BMI calculations, we evaluated five LD reference panels and three sets of GWAS summary statistics of varying sample size and ancestry. PRS_BMI performance increased for both African and European ancestry individuals using cross-ancestry GWAS summary statistics compared to European-only summary statistics (6.3% and 3.7% relative R² increase, respectively, p_African=0.038, p_European=6.26x10^-4). The effects of LD reference panels were more pronounced in African ancestry study datasets. PRS_BMI performance degraded in children; R² was less than half of teenagers or adults. The effect of GWAS summary statistics sample size was small when modeled with the other factors. Additionally, the potential of using a PRS generated for one trait to predict risk for comorbid diseases is not well understood especially in the context of cross-ancestry analyses – we explored clinical comorbidities from the electronic health record associated with PRS_BMI and identified significant associations with type 2 diabetes and coronary atherosclerosis. In summary, this study quantifies the effects that ancestry, GWAS summary statistic sample size, and LD reference panel have on PRS performance, especially in cross-ancestry and age-specific analyses.

The Diversity and Disparity in Biomedical Informatics (DDBI) workshop will be focused on complementary and critical issues concerned with enhancing diversity in the informatics workforce as well as diversity in patient cohorts. According to the National Institute of Minority Health and Health Disparities (NIMHD) at the NIH, diversity refers to the inclusion of the following traditionally underrepresented groups: African Americans/Blacks, Asians (>30 countries), American Indian or Alaska Native, Native Hawaiian or Other Pacific Islander, Latino or Hispanic (20 countries). Gender, culture, and socioeconomic status are also important dimensions of diversity, which may define some underrepresented groups. The under-representation of specific groups in both the biomedical informatics workforce as well as in the patient-derived data that is being used for research purposes has contributed to an ongoing disparity; these groups have not experienced equity in contributing to or benefiting from advancements in informatics research. This workshop will highlight innovative efforts to increase the pool of minority informaticians and discuss examples of informatics research that addresses the health concerns that impact minority populations. This workshop topics will provide insight into overcoming pipeline issues in the development of minority informaticians while emphasizing the importance of minority participation in health related research. The DDBI workshop will occur in two parts. Part I will discuss specific minority health & health disparities research topics and Part II will cover discussions related to overcoming pipeline issues in the training of minority informaticians.