arXiv subjects

Brian Caffo

Publications and source records attributed to Brian Caffo.

At least 19 recordsLinked to original sources

arXiv2026

Classifier Pooling for Modern Ordinal Classification

Ordinal data is widely prevalent in clinical and other domains, yet there is a lack of both modern, machine-learning based methods and publicly available software to address it. In this paper, we present a model-agnostic method of ordinal classification, which can apply any non-ordinal classification method in an ordinal fashion. We also provide an open-source implementation of these algorithms, in the form of a Python package. We apply these models on multiple real-world datasets to show their performance across domains. We show that they often outperform non-ordinal classification methods, especially when the number of datapoints is relatively small or when there are many classes of outcomes. This work, including the developed software, facilitates the use of modern, more powerful machine learning algorithms to handle ordinal data.

cs.LG↗

arXiv2026

Adaptive Label Error Detection: A Bayesian Approach to Mislabeled Data Detection

Machine learning classification systems are susceptible to poor performance when trained with incorrect ground truth labels, even when data is well-curated by expert annotators. As machine learning becomes more widespread, it is increasingly imperative to identify and correct mislabeling to develop more powerful models. In this work, we motivate and describe Adaptive Label Error Detection (ALED), a novel method of detecting mislabeling. ALED extracts an intermediate feature space from a deep convolutional neural network, denoises the features, models the reduced manifold of each class with a multidimensional Gaussian distribution, and performs a simple likelihood ratio test to identify mislabeled samples. We show that ALED has markedly increased sensitivity, without compromising precision, compared to established label error detection methods, on multiple medical imaging datasets. We demonstrate an example where fine-tuning a neural network on corrected data results in a 33.8% decrease in test set errors, providing strong benefits to end users. The ALED detector is deployed in the Python package statlab.

cs.LG↗

arXiv2025

BrainMT: A Hybrid Mamba-Transformer Architecture for Modeling Long-Range Dependencies in Functional MRI Data

Recent advances in deep learning have made it possible to predict phenotypic measures directly from functional magnetic resonance imaging (fMRI) brain volumes, sparking significant interest in the neuroimaging community. However, existing approaches, primarily based on convolutional neural networks or transformer architectures, often struggle to model the complex relationships inherent in fMRI data, limited by their inability to capture long-range spatial and temporal dependencies. To overcome these shortcomings, we introduce BrainMT, a novel hybrid framework designed to efficiently learn and integrate long-range spatiotemporal attributes in fMRI data. Our framework operates in two stages: (1) a bidirectional Mamba block with a temporal-first scanning mechanism to capture global temporal interactions in a computationally efficient manner; and (2) a transformer block leveraging self-attention to model global spatial relationships across the deep features processed by the Mamba block. Extensive experiments on two large-scale public datasets, UKBioBank and the Human Connectome Project, demonstrate that BrainMT achieves state-of-the-art performance on both classification (sex prediction) and regression (cognitive intelligence prediction) tasks, outperforming existing methods by a significant margin. Our code and implementation details will be made publicly available at this https://github.com/arunkumar-kannan/BrainMT-fMRI

cs.CV↗

arXiv2025

Cross-Attention Fusion of MRI and Jacobian Maps for Alzheimer's Disease Diagnosis

Early diagnosis of Alzheimer's disease (AD) is critical for intervention before irreversible neurodegeneration occurs. Structural MRI (sMRI) is widely used for AD diagnosis, but conventional deep learning approaches primarily rely on intensity-based features, which require large datasets to capture subtle structural changes. Jacobian determinant maps (JSM) provide complementary information by encoding localized brain deformations, yet existing multimodal fusion strategies fail to fully integrate these features with sMRI. We propose a cross-attention fusion framework to model the intrinsic relationship between sMRI intensity and JSM-derived deformations for AD classification. Using the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset, we compare cross-attention, pairwise self-attention, and bottleneck attention with four pre-trained 3D image encoders. Cross-attention fusion achieves superior performance, with mean ROC-AUC scores of 0.903 (+/-0.033) for AD vs. cognitively normal (CN) and 0.692 (+/-0.061) for mild cognitive impairment (MCI) vs. CN. Despite its strong performance, our model remains highly efficient, with only 1.56 million parameters--over 40 times fewer than ResNet-34 (63M) and Swin UNETR (61.98M). These findings demonstrate the potential of cross-attention fusion for improving AD diagnosis while maintaining computational efficiency.

eess.IV↗

arXiv2023

Evidential Uncertainty Quantification: A Variance-Based Perspective

Uncertainty quantification of deep neural networks has become an active field of research and plays a crucial role in various downstream tasks such as active learning. Recent advances in evidential deep learning shed light on the direct quantification of aleatoric and epistemic uncertainties with a single forward pass of the model. Most traditional approaches adopt an entropy-based method to derive evidential uncertainty in classification, quantifying uncertainty at the sample level. However, the variance-based method that has been widely applied in regression problems is seldom used in the classification setting. In this work, we adapt the variance-based approach from regression to classification, quantifying classification uncertainty at the class level. The variance decomposition technique in regression is extended to class covariance decomposition in classification based on the law of total covariance, and the class correlation is also derived from the covariance. Experiments on cross-domain datasets are conducted to illustrate that the variance-based approach not only results in similar accuracy as the entropy-based one in active domain adaptation but also brings information about class-wise uncertainties as well as between-class correlations. The code is available at https://github.com/KerryDRX/EvidentialADA. This alternative means of evidential uncertainty quantification will give researchers more options when class uncertainties and correlations are important in their applications.

cs.LG↗

arXiv2023

Applications of Sequential Learning for Medical Image Classification

Purpose: The aim of this work is to develop a neural network training framework for continual training of small amounts of medical imaging data and create heuristics to assess training in the absence of a hold-out validation or test set. Materials and Methods: We formulated a retrospective sequential learning approach that would train and consistently update a model on mini-batches of medical images over time. We address problems that impede sequential learning such as overfitting, catastrophic forgetting, and concept drift through PyTorch convolutional neural networks (CNN) and publicly available Medical MNIST and NIH Chest X-Ray imaging datasets. We begin by comparing two methods for a sequentially trained CNN with and without base pre-training. We then transition to two methods of unique training and validation data recruitment to estimate full information extraction without overfitting. Lastly, we consider an example of real-life data that shows how our approach would see mainstream research implementation. Results: For the first experiment, both approaches successfully reach a ~95% accuracy threshold, although the short pre-training step enables sequential accuracy to plateau in fewer steps. The second experiment comparing two methods showed better performance with the second method which crosses the ~90% accuracy threshold much sooner. The final experiment showed a slight advantage with a pre-training step that allows the CNN to cross ~60% threshold much sooner than without pre-training. Conclusion: We have displayed sequential learning as a serviceable multi-classification technique statistically comparable to traditional CNNs that can acquire data in small increments feasible for clinically realistic scenarios.

eess.IV↗

arXiv2023

Learning sources of variability from high-dimensional observational studies

Causal inference studies whether the presence of a variable influences an observed outcome. As measured by quantities such as the "average treatment effect," this paradigm is employed across numerous biological fields, from vaccine and drug development to policy interventions. Unfortunately, the majority of these methods are often limited to univariate outcomes. Our work generalizes causal estimands to outcomes with any number of dimensions or any measurable space, and formulates traditional causal estimands for nominal variables as causal discrepancy tests. We propose a simple technique for adjusting universally consistent conditional independence tests and prove that these tests are universally consistent causal discrepancy tests. Numerical experiments illustrate that our method, Causal CDcorr, leads to improvements in both finite sample validity and power when compared to existing strategies. Our methods are all open source and available at github.com/ebridge2/cdcorr.

stat.ME↗

arXiv2023

Direct Bayesian Regression for Distribution-valued Covariates

In this manuscript, we study the problem of scalar-on-distribution regression; that is, instances where subject-specific distributions or densities, or in practice, repeated measures from those distributions, are the covariates related to a scalar outcome via a regression model. We propose a direct regression for such distribution-valued covariates that circumvents estimating subject-specific densities and directly uses the observed repeated measures as covariates. The model is invariant to any transformation or ordering of the repeated measures. Endowing the regression function with a Gaussian Process prior, we obtain closed form or conjugate Bayesian inference. Our method subsumes the standard Bayesian non-parametric regression using Gaussian Processes as a special case. Theoretically, we show that the method can achieve an optimal estimation error bound. To our knowledge, this is the first theoretical study on Bayesian regression using distribution-valued covariates. Through simulation studies and analysis of activity count dataset, we demonstrate that our method performs better than approaches that require an intermediate density estimation step.

stat.ME↗

arXiv2022

The Multiple Subnetwork Hypothesis: Enabling Multidomain Learning by Isolating Task-Specific Subnetworks in Feedforward Neural Networks

Neural networks have seen an explosion of usage and research in the past decade, particularly within the domains of computer vision and natural language processing. However, only recently have advancements in neural networks yielded performance improvements beyond narrow applications and translated to expanded multitask models capable of generalizing across multiple data types and modalities. Simultaneously, it has been shown that neural networks are overparameterized to a high degree, and pruning techniques have proved capable of significantly reducing the number of active weights within the network while largely preserving performance. In this work, we identify a methodology and network representational structure which allows a pruned network to employ previously unused weights to learn subsequent tasks. We employ these methodologies on well-known benchmarking datasets for testing purposes and show that networks trained using our approaches are able to learn multiple tasks, which may be related or unrelated, in parallel or in sequence without sacrificing performance on any task or exhibiting catastrophic forgetting.

cs.LG↗

arXiv2022

Prospective Learning: Principled Extrapolation to the Future

Learning is a process which can update decision rules, based on past experience, such that future performance improves. Traditionally, machine learning is often evaluated under the assumption that the future will be identical to the past in distribution or change adversarially. But these assumptions can be either too optimistic or pessimistic for many problems in the real world. Real world scenarios evolve over multiple spatiotemporal scales with partially predictable dynamics. Here we reformulate the learning problem to one that centers around this idea of dynamic futures that are partially learnable. We conjecture that certain sequences of tasks are not retrospectively learnable (in which the data distribution is fixed), but are prospectively learnable (in which distributions may be dynamic), suggesting that prospective learning is more difficult in kind than retrospective learning. We argue that prospective learning more accurately characterizes many real world problems that (1) currently stymie existing artificial intelligence solutions and/or (2) lack adequate explanations for how natural intelligences solve them. Thus, studying prospective learning will lead to deeper insights and solutions to currently vexing challenges in both natural and artificial intelligences.

cs.LG↗

arXiv2020

Statistical Analysis of Data Repeatability Measures

The advent of modern data collection and processing techniques has seen the size, scale, and complexity of data grow exponentially. A seminal step in leveraging these rich datasets for downstream inference is understanding the characteristics of the data which are repeatable -- the aspects of the data that are able to be identified under a duplicated analysis. Conflictingly, the utility of traditional repeatability measures, such as the intraclass correlation coefficient, under these settings is limited. In recent work, novel data repeatability measures have been introduced in the context where a set of subjects are measured twice or more, including: fingerprinting, rank sums, and generalizations of the intraclass correlation coefficient. However, the relationships between, and the best practices among these measures remains largely unknown. In this manuscript, we formalize a novel repeatability measure, discriminability. We show that it is deterministically linked with the correlation coefficient under univariate random effect models, and has desired property of optimal accuracy for inferential tasks using multivariate measurements. Additionally, we overview and systematically compare repeatability statistics using both theoretical results and simulations. We show that the rank sum statistic is deterministically linked to a consistent estimator of discriminability. The power of permutation tests derived from these measures are compared numerically under Gaussian and non-Gaussian settings, with and without simulated batch effects. Motivated by both theoretical and empirical results, we provide methodological recommendations for each benchmark setting to serve as a resource for future analyses. We believe these recommendations will play an important role towards improving repeatability in fields such as functional magnetic resonance imaging, genomics, pharmacology, and more.

stat.AP↗

arXiv2020

Alpha-1 adrenergic receptor antagonists to prevent hyperinflammation and death from lower respiratory tract infection

In severe viral pneumonia, including Coronavirus disease 2019 (COVID-19), the viral replication phase is often followed by hyperinflammation, which can lead to acute respiratory distress syndrome, multi-organ failure, and death. We previously demonstrated that alpha-1 adrenergic receptor ($\alpha_1$-AR) antagonists can prevent hyperinflammation and death in mice. Here, we conducted retrospective analyses in two cohorts of patients with acute respiratory distress (ARD, n=18,547) and three cohorts with pneumonia (n=400,907). Federated across two ARD cohorts, we find that patients exposed to $\alpha_1$-AR antagonists, as compared to unexposed patients, had a 34% relative risk reduction for mechanical ventilation and death (OR=0.70, p=0.021). We replicated these methods on three pneumonia cohorts, all with similar effects on both outcomes. All results were robust to sensitivity analyses. These results highlight the urgent need for prospective trials testing whether prophylactic use of $\alpha_1$-AR antagonists ameliorates lower respiratory tract infection-associated hyperinflammation and death, as observed in COVID-19.

q-bio.TO↗

arXiv2018

Functional Mediation Analysis with an Application to Functional Magnetic Resonance Imaging Data

Causal mediation analysis is widely utilized to separate the causal effect of treatment into its direct effect on the outcome and its indirect effect through an intermediate variable (the mediator). In this study we introduce a functional mediation analysis framework in which the three key variables, the treatment, mediator, and outcome, are all continuous functions. With functional measures, causal assumptions and interpretations are not immediately well-defined. Motivated by a functional magnetic resonance imaging (fMRI) study, we propose two functional mediation models based on the influence of the mediator: (1) a concurrent mediation model and (2) a historical mediation model. We further discuss causal assumptions, and elucidate causal interpretations. Our proposed models enable the estimation of individual causal effect curves, where both the direct and indirect effects vary across time. Applied to a task-based fMRI study, we illustrate how our functional mediation framework provides a new perspective for studying dynamic brain connectivity. The R package cfma is available on CRAN.

stat.AP↗

arXiv2016

PCA leverage: outlier detection for high-dimensional functional magnetic resonance imaging data

Outlier detection for high-dimensional (HD) data is a popular topic in modern statistical research. However, one source of HD data that has received relatively little attention is functional magnetic resonance images (fMRI), which consists of hundreds of thousands of measurements sampled at hundreds of time points. At a time when the availability of fMRI data is rapidly growing---primarily through large, publicly available grassroots datasets---automated quality control and outlier detection methods are greatly needed. We propose PCA leverage and demonstrate how it can be used to identify outlying time points in an fMRI run. Furthermore, PCA leverage is a measure of the influence of each observation on the estimation of principal components, which are often of interest in fMRI data. We also propose an alternative measure, PCA robust distance, which is less sensitive to outliers and has controllable statistical properties. The proposed methods are validated through simulation studies and are shown to be highly accurate. We also conduct a reliability study using resting-state fMRI data from the Autism Brain Imaging Data Exchange and find that removal of outliers using the proposed methods results in more reliable estimation of subject-level resting-state networks using ICA.

stat.ME↗

arXiv2016

Stability and Localization of inter-individual differences in functional connectivity

Much recent attention has been paid to quantifying anatomic and functional neuroimaging on the individual subject level. For optimal individual subject characterization, specific acquisition and analysis features need to be identified that maximize inter-individual variability while concomitantly minimizing intra-subject variability. Here we develop a non-parametric statistical metric that quantifies the degree to which a parameter set allows this individual subject differentiation. We apply this metric to analyzing publicly available test-retest resting-state fMRI (rs-fMRI) data sets. We find that for the question of maximizing individual differentiation, there is a relative tradeoff between increasing sampling through increased sampling frequency or increased acquisition time; that for the sizes of the interrogated data sets, only 4-5 min of acquisition time is necessary to perfectly differentiate each subject; and that brain regions that most contribute to individuals unique characterization lie in association cortices thought to contribute to higher cognitive function. These findings may guide optimal rs-fMRI experiment design and may aid elucidation of the neural bases for subject-to-subject differences.

stat.AP↗

arXiv2015

Improving Reliability of Subject-Level Resting-State fMRI Parcellation with Shrinkage Estimators

A recent interest in resting state functional magnetic resonance imaging (rsfMRI) lies in subdividing the human brain into anatomically and functionally distinct regions of interest. For example, brain parcellation is often used for defining the network nodes in connectivity studies. While inference has traditionally been performed on group-level data, there is a growing interest in parcellating single subject data. However, this is difficult due to the low signal-to-noise ratio of rsfMRI data, combined with typically short scan lengths. A large number of brain parcellation approaches employ clustering, which begins with a measure of similarity or distance between voxels. The goal of this work is to improve the reproducibility of single-subject parcellation using shrinkage estimators of such measures, allowing the noisy subject-specific estimator to "borrow strength" in a principled manner from a larger population of subjects. We present several empirical Bayes shrinkage estimators and outline methods for shrinkage when multiple scans are not available for each subject. We perform shrinkage on raw intervoxel correlation estimates and use both raw and shrinkage estimates to produce parcellations by performing clustering on the voxels. Our proposed method is agnostic to the choice of clustering method and can be used as a pre-processing step for any clustering algorithm. Using two datasets---a simulated dataset where the true parcellation is known and is subject-specific and a test-retest dataset consisting of two 7-minute rsfMRI scans from 20 subjects---we show that parcellations produced from shrinkage correlation estimates have higher reliability and validity than those produced from raw estimates. Application to test-retest data shows that using shrinkage estimators increases the reproducibility of subject-specific parcellations of the motor cortex by up to 30%.

stat.ME↗

arXiv2014

On the Impact of Dimension Reduction on Graphical Structures

Statisticians and quantitative neuroscientists have actively promoted the use of independence relationships for investigating brain networks, genomic networks, and other measurement technologies. Estimation of these graphs depends on two steps. First is a feature extraction by summarizing measurements within a parcellation, regional or set definition to create nodes. Secondly, these summaries are then used to create a graph representing relationships of interest. In this manuscript we study the impact of dimension reduction on graphs that describe different notions of relations among a set of random variables. We are particularly interested in undirected graphs that capture the random variables' independence and conditional independence relations. A dimension reduction procedure can be any mapping from high dimensional spaces to low dimensional spaces. We exploit a general framework for modeling the raw data and advocate that in estimating the undirected graphs, any acceptable dimension reduction procedure should be a graph-homotopic mapping, i.e., the graphical structure of the data after dimension reduction should inherit the main characteristics of the graphical structure of the raw data. We show that, in terms of inferring undirected graphs that characterize the conditional independence relations among random variables, many dimension reduction procedures, such as the mean, median, or principal components, cannot be theoretically guaranteed to be a graph-homotopic mapping. The implications of this work are broad. In the most charitable setting for researchers, where the correct node definition is known, graphical relationships can be contaminated merely via the dimension reduction. The manuscript ends with a concrete example, characterizing a subset of graphical structures such that the dimension reduction procedure using the principal components can be a graph-homotopic mapping.

stat.ME↗

arXiv2014

Joint Estimation of Multiple Graphical Models from High Dimensional Time Series

In this manuscript we consider the problem of jointly estimating multiple graphical models in high dimensions. We assume that the data are collected from n subjects, each of which consists of T possibly dependent observations. The graphical models of subjects vary, but are assumed to change smoothly corresponding to a measure of closeness between subjects. We propose a kernel based method for jointly estimating all graphical models. Theoretically, under a double asymptotic framework, where both (T,n) and the dimension d can increase, we provide the explicit rate of convergence in parameter estimation. It characterizes the strength one can borrow across different individuals and impact of data dependence on parameter estimation. Empirically, experiments on both synthetic and real resting state functional magnetic resonance imaging (rs-fMRI) data illustrate the effectiveness of the proposed method.

stat.ML↗