Predicting functional impairments with lesion-derived disconnectome mapping: Validation in stroke patients with motor deficits

Focal structural damage to white matter tracts can result in functional deficits in stroke patients. Traditional voxel-based lesion-symptom mapping is commonly used to localize brain structures linked to neurological deficits. Emerging evidence suggests that the impact of structural focal damage may extend beyond immediate lesion sites. In this study, we present a disconnectome mapping approach based on support vector regression (SVR) to identify brain structures and white matter pathways associated with functional deficits in stroke patients. For clinical validation, we utilized imaging data from 340 stroke patients exhibiting motor deficits. A disconnectome map was initially derived from lesions for each patient. Bootstrap sampling was then employed to balance the sample size between a minority group of patients exhibiting right or left motor deficits and those without deficits. Subsequently, SVR analysis was used to identify voxels associated with motor deficits (p < .005). Our disconnectome-based analysis significantly outperformed alternative lesion-symptom approaches in identifying major white matter pathways within the corticospinal tracts associated with upper-lower limb motor deficits. Bootstrapping significantly increased the sensitivity (80%-87%) for identifying patients with motor deficits, with a minimum lesion size of 32 and 235 mm3 for the right and left motor deficit, respectively. Overall, the lesion-based methods achieved lower sensitivities compared with those based on disconnection maps. The primary contribution of our approach lies in introducing a bootstrapped disconnectome-based mapping approach to identify lesion-derived white matter disconnections associated with functional deficits, particularly efficient in handling imbalanced data.

An artificial immune system with bootstrap sampling for the diagnosis of recurrent endometrial cancers

Endometrial cancer is one of the most common gynecological malignancies in developed countries. The prevention of the recurrence of endometrial cancer has always been a clinical challenge. Endometrial cancer is asymptomatic in the early stage, and there remains a lack of time-series correlation patterns of clinical pathway transfer, recurrence, and treatment. In this study, the artificial immune system (AIS) combined with bootstrap sampling was compared with other machine learning techniques, which included both supervised and unsupervised learning categories. The back propagation neural network, support vector machine (SVM) with a radial basis function kernel, fuzzy c-means, and ant k-means were compared with the proposed method to verify the sensitivity and specificity of the datasets, and the important factors of recurrent endometrial cancer were predicted. In the unsupervised learning algorithms, the AIS algorithm had the highest accuracy (83.35%), sensitivity (77.35%), and specificity (92.31%); in supervised learning algorithms, the SVM algorithm had the highest accuracy (97.51%), sensitivity (95.02%), and specificity (99.29%). The results of our study showed that histology and chemotherapy are important factors affecting the prediction of recurrence. Finally, behavior code and radiotherapy for recurrent endometrial cancer are important factors for future adjuvant treatment.

Ensemble Learning with Stochastic Configuration Network for Noisy Optical Fiber Vibration Signal Recognition

Optical fiber pre-warning systems (OFPS) based on Φ-OTDR are applied to many different scenarios such as oil and gas pipeline protection. The recognition of fiber vibration signals is one of the most important parts of this system. According to the characteristics of small sample set, we choose stochastic configuration network (SCN) for recognition. However, due to the interference of environmental and mechanical noise, the recognition effect of vibration signals will be affected. In order to study the effect of noise on signal recognition performance, we recognize noisy optical fiber vibration signals, which superimposed analog white Gaussian noise, white uniform noise, Rayleigh distributed noise, and exponentially distributed noise. Meanwhile, bootstrap sampling (bagging) and AdaBoost ensemble learning methods are combined with original SCN, and Bootstrap-SCN, AdaBoost-SCN, and AdaBoost-Bootstrap-SCN are proposed and compared for noisy signals recognition. Results show that: (1) the recognition rates of two classifiers combined with AdaBoost are higher than the other two methods over the entire noise range; (2) the recognition for noisy signals of AdaBoost-Bootstrap-SCN is better than other methods in recognition of noisy signals.

Assessing genomic prediction accuracy for Holstein sires using bootstrap aggregation sampling and leave-one-out cross validation

Since the introduction of genome-enabled prediction for dairy cattle in 2009, genomic selection has markedly changed many aspects of the dairy genetics industry and enhanced the rate of response to selection for most economically important traits. Young dairy bulls are genotyped to obtain their genomic predicted transmitting ability (GPTA) and reliability (REL) values. These GPTA are a main factor in most purchasing, marketing, and culling decisions until bulls reach 5 yr of age and their milk-recorded offspring become available. At that time, daughter yield deviations (DYD) can be compared with the GPTA computed several years earlier. For most bulls, the DYD align well with the initial predictions. However, for some bulls, the difference between DYD and corresponding GPTA is quite large, and published REL are of limited value in identifying such bulls. A method of bootstrap aggregation sampling (bagging) using genomic BLUP (GBLUP) was applied to predict the GPTA of 2,963, 2,963, and 2,803 young Holstein bulls for protein yield, somatic cell score, and daughter pregnancy rate (DPR), respectively. For each trait, 50 bootstrap samples from a reference population comprising 2011 DYD of 8,610, 8,405, and 7,945 older Holstein bulls were used. Leave-one-out cross validation was also performed to assess prediction accuracy when removing specific bulls from the reference population. The main objectives of this study were (1) to assess the extent to which current REL values and alternative measures of variability, such as the bootstrap standard deviation (SD) of predictions, could detect bulls whose daughter performance deviates significantly from early genomic predictions, and (2) to identify factors associated with the reference population that inform about inaccurate genomic predictions. The SD of bootstrap predictions was a mildly useful metric for identifying bulls whose future daughter performance may deviate significantly from early GPTA for protein and DPR. Leave-one-out cross validation allowed us to identify groups of reference population bulls that were influential on other reference population bulls for protein yield and observe their effects on predictions of testing set bulls, as a whole and individually.

Variable selection methods for developing a biomarker panel for prediction of dengue hemorrhagic fever

BACKGROUND

The choice of selection methods to identify important variables for binary classification modeling is critical to produce stable models that are interpretable, that generate accurate predictions and have minimum bias. This work is motivated by data on clinical and laboratory features of severe dengue infections (dengue hemorrhagic fever, DHF) obtained from 51 individuals enrolled in a prospective observational study of acute human dengue infections.

RESULTS

We carry out a comprehensive performance comparison using several classification models for DHF over the dengue data set. We compared variable selection results by Multivariate Adaptive Regression Splines, Learning Ensemble, Random Forest, Bayesian Moving Averaging, Stochastic Search Variable Selection, and Generalized Regularized Logistics Regression. Model averaging methods (bagging, boosting and ensemble learners) have higher accuracy, but the generalized regularized regression model has the highest predictive power because the linearity assumptions of candidate predictors are strongly satisfied via deviance chi-square testing procedures. Bootstrapping applications for evaluating predictive regression coefficients in regularized regression model are performed.

CONCLUSIONS

Feature reduction methods introduce inherent biases and therefore are data-type dependent. We propose that these limitations can be overcome using an exhaustive approach for searching feature space. Using this approach, our results suggest that IL-10, platelet and lymphocyte counts are the major features for predicting dengue DHF on the basis of blood chemistries and cytokine measurements.

A bootstrap approach to estimating power for linkage heterogeneity

We examined the power of detecting linkage heterogeneity when the null hypothesis is that all families are linked to one locus (A) and the two alternative hypotheses are either 1) a proportion of the families are linked to locus A and the remaining families are linked to a second locus B or 2) a proportion of the families are linked to locus A or B and a third proportion of the families are unlinked to either locus. The power of detecting linkage heterogeneity is estimated for various proportions of families linked to loci A, B or unlinked to either locus (sampling under the alternative hypothesis). To estimate the significance level, the data set is sampled under the null hypothesis. For sampling under both hypotheses, a bootstrap approach is employed, sampling the simulated pedigrees with replacement. The power to detect linkage heterogeneity is strongest when the recombination fraction is 0 and equal proportions of the families are linked to loci A and B. The power decreases as the recombination fraction increases, the proportion of unlinked families increases and the disparity between the proportion of the families linked to either locus A or B increases. In the data set of 32 Duke Familial Alzheimer Disease families, when equal proportions of families are linked to loci A and B, the power to detect linkage heterogeneity is 0.94 using a likelihood ratio criterion of 10:1. The p value that corresponds to the likelihood ratio of 10:1 is estimated as 0.013 with a 95% confidence interval for p ranging from 0.012 to 0.014.