arXiv subjects

Duy Duong-Tran

Publications and source records attributed to Duy Duong-Tran.

At least 19 recordsLinked to original sources

arXiv2026

Interpretable Alzheimer's Diagnosis via Multimodal Fusion of Regional Brain Experts

Accurate and early diagnosis of Alzheimer's disease (AD) is critical for effective intervention and requires integrating complementary information from multimodal neuroimaging data. However, conventional fusion approaches often rely on simple concatenation of features, which cannot adaptively balance the contributions of biomarkers such as amyloid PET and MRI across brain regions. In this work, we propose MREF-AD, a Multimodal Regional Expert Fusion model for AD diagnosis. It is a Mixture-of-Experts (MoE) framework that models mesoscopic brain regions within each modality as independent experts and employs a gating network to learn subject-specific fusion weights. Utilizing tabular neuroimaging and demographic information from the Alzheimer's Disease Neuroimaging Initiative (ADNI), MREF-AD achieves competitive performance over strong classic and deep baselines while providing interpretable, modality- and region-level insight into how structural and molecular imaging jointly contribute to AD diagnosis. The source code is available at https://github.com/PennShenLab/mref-ad.

cs.LG↗

arXiv2026

Tabular LLMs for Interpretable Few-Shot Alzheimer's Disease Prediction with Multimodal Biomedical Data

Accurate diagnosis of Alzheimer's disease (AD) requires handling tabular biomarker data, yet such data are often small and incomplete, where deep learning models frequently fail to outperform classical methods. Pretrained large language models (LLMs) offer few-shot generalization, structured reasoning, and interpretable outputs, providing a powerful paradigm shift for clinical prediction. We propose TAP-GPT Tabular Alzheimer's Prediction GPT, a domain-adapted tabular LLM framework built on TableGPT2 and fine-tuned for few-shot AD classification using tabular prompts rather than plain texts. We evaluate TAP-GPT across four ADNI-derived datasets, including QT-PAD biomarkers and region-level structural MRI, amyloid PET, and tau PET for binary AD classification. Across multimodal and unimodal settings, TAP-GPT improves upon its backbone models and outperforms traditional machine learning baselines in the few-shot setting while remaining competitive with state-of-the-art general-purpose LLMs. We show that feature selection mitigates degradation in high-dimensional inputs and that TAP-GPT maintains stable performance under simulated and real-world missingness without imputation. Additionally, TAP-GPT produces structured, modality-aware reasoning aligned with established AD biology and shows greater stability under self-reflection, supporting its use in iterative multi-agent systems. To our knowledge, this is the first systematic application of a tabular-specialized LLM to multimodal biomarker-based AD prediction, demonstrating that such pretrained models can effectively address structured clinical prediction tasks and laying the foundation for tabular LLM-driven multi-agent clinical decision-support systems. The source code is publicly available on GitHub: https://github.com/sophie-kearney/TAP-GPT.

cs.CL↗

arXiv2025

Enabling Few-Shot Alzheimer's Disease Diagnosis on Biomarker Data with Tabular LLMs

Early and accurate diagnosis of Alzheimer's disease (AD), a complex neurodegenerative disorder, requires analysis of heterogeneous biomarkers (e.g., neuroimaging, genetic risk factors, cognitive tests, and cerebrospinal fluid proteins) typically represented in a tabular format. With flexible few-shot reasoning, multimodal integration, and natural-language-based interpretability, large language models (LLMs) offer unprecedented opportunities for prediction with structured biomedical data. We propose a novel framework called TAP-GPT, Tabular Alzheimer's Prediction GPT, that adapts TableGPT2, a multimodal tabular-specialized LLM originally developed for business intelligence tasks, for AD diagnosis using structured biomarker data with small sample sizes. Our approach constructs few-shot tabular prompts using in-context learning examples from structured biomedical data and finetunes TableGPT2 using the parameter-efficient qLoRA adaption for a clinical binary classification task of AD or cognitively normal (CN). The TAP-GPT framework harnesses the powerful tabular understanding ability of TableGPT2 and the encoded prior knowledge of LLMs to outperform more advanced general-purpose LLMs and a tabular foundation model (TFM) developed for prediction tasks. To our knowledge, this is the first application of LLMs to the prediction task using tabular biomarker data, paving the way for future LLM-driven multi-agent frameworks in biomedical informatics.

cs.CL↗

arXiv2025

Alpha-Z divergence unveils further distinct phenotypic traits of human brain connectivity fingerprint

The accurate identification of individuals from functional connectomes (FCs) is critical for advancing individualized assessments in neuropsychiatric research. Traditional methods, such as Pearson's correlation, have limitations in capturing the complex, non-Euclidean geometry of FC data, leading to suboptimal performance in identification performance. Recent developments have introduced geodesic distance as a more robust metric; however, its performance is highly sensitive to regularization choices, which vary by spatial scale and task condition. To address these challenges, we propose a novel divergence-based distance metric, the Alpha-Z Bures-Wasserstein divergence, which provides a more flexible and geometry-aware framework for FC comparison. Unlike prior methods, our approach does not require meticulous parameter tuning and maintains strong identification performance across multiple task conditions and spatial resolutions. We evaluate our approach against both traditional (e.g., Euclidean, Pearson) and state-of-the-art manifold-based distances (e.g., affine-invariant, log-Euclidean, Bures-Wasserstein), and systematically investigate how varying regularization strengths affect geodesic distance performance on the Human Connectome Project dataset. Our results show that the proposed method significantly improves identification rates over traditional and geodesic distances, particularly when optimized regularization is applied, and especially in high-dimensional settings where matrix rank deficiencies degrade existing metrics. We further validate its generalizability across resting-state and task-based fMRI, using multiple parcellation schemes. These findings suggest that the new divergence provides a more reliable and generalizable framework for functional connectivity analysis, offering enhanced sensitivity in linking FC patterns to cognitive and behavioral outcomes.

q-bio.NC↗

arXiv2025

Data-Driven Stabilization of Unknown Linear-Threshold Network Dynamics

This paper studies the data-driven control of unknown linear-threshold network dynamics to stabilize the state to a reference value. We consider two types of controllers: (i) a state feedback controller with feed-forward reference input and (ii) an augmented feedback controller with error integration. The first controller features a simpler structure and is easier to design, while the second offers improved performance in the presence of system parameter changes and disturbances. Our design strategy employs state-input datasets to construct data-based representations of the closed-loop dynamics. Since these representations involve linear threshold functions, we rewrite them as switched linear systems, and formulate the design problem as that of finding a common controller for all the resulting modes. This gives rise to a set of linear matrix inequalities (LMIs) whose solutions corresponds to the controller gain matrices. We analyze the computational complexity of solving the LMIs and propose a simplified, sufficient set of conditions that scales linearly with the system state. Simulations on two case studies involving regulation of firing rate dynamics in rodent brains and of arousal level dynamics in humans demonstrate the effectiveness of the controller designs.

eess.SY↗

arXiv2025

Reconstructing Brain Causal Dynamics for Subject and Task Fingerprints using fMRI Time-series Data

Purpose: Recently, there has been a revived interest in system neuroscience causation models, driven by their unique capability to unravel complex relationships in multi-scale brain networks. In this paper, we present a novel method that leverages causal dynamics to achieve effective fMRI-based subject and task fingerprinting. Methods: By applying an implicit-explicit discretization scheme, we develop a two-timescale linear state-space model. Through data-driven identification of its parameters, the model captures causal signatures, including directed interactions among brain regions from a spatial perspective, and disentangled fast and slow dynamic modes of brain activity from a temporal perspective. These causal signatures are then integrated with: (i) a modal decomposition and projection method for model-based subject identification, and (ii) a Graph Neural Network (GNN) framework for learning-based task classification. Furthermore, we introduce the concept of the brain reachability landscape as a novel visualization tool, which quantitatively characterizes the maximum possible activation levels of brain regions under various fMRI tasks. Results: We evaluate the proposed approach using the Human Connectome Project dataset and demonstrate its advantage over non-causality-based methods. The obtained causal signatures are visualized and demonstrate clear biological relevance with established understandings of brain function. Conclusion: We verified the feasibility and effectiveness of utilizing brain causal signatures for subject and task fingerprinting. Additionally, our work paves the way for further studies on causal fingerprints with potential applications in both healthy controls and neurodegenerative diseases.

eess.SY↗

arXiv2025

Fair CCA for Fair Representation Learning: An ADNI Study

Canonical correlation analysis (CCA) is a technique for finding correlations between different data modalities and learning low-dimensional representations. As fairness becomes crucial in machine learning, fair CCA has gained attention. However, previous approaches often overlook the impact on downstream classification tasks, limiting applicability. We propose a novel fair CCA method for fair representation learning, ensuring the projected features are independent of sensitive attributes, thus enhancing fairness without compromising accuracy. We validate our method on synthetic data and real-world data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), demonstrating its ability to maintain high correlation analysis performance while improving fairness in classification tasks. Our work enables fair machine learning in neuroimaging studies where unbiased analysis is essential. Code is available in https://github.com/ZhanliangAaronWang/FR-CCA-ADNI.

cs.LG↗

arXiv2025

From Promising Capability to Pervasive Bias: Assessing Large Language Models for Emergency Department Triage

Large Language Models (LLMs) have shown promise in clinical decision support, yet their application to triage remains underexplored. We systematically investigate the capabilities of LLMs in emergency department triage through two key dimensions: (1) robustness to distribution shifts and missing data, and (2) counterfactual analysis of intersectional biases across sex and race. We assess multiple LLM-based approaches, ranging from continued pre-training to in-context learning, as well as machine learning approaches. Our results indicate that LLMs exhibit superior robustness, and we investigate the key factors contributing to the promising LLM-based approaches. Furthermore, in this setting, we identify gaps in LLM preferences that emerge in particular intersections of sex and race. LLMs generally exhibit sex-based differences, but they are most pronounced in certain racial groups. These findings suggest that LLMs encode demographic preferences that may emerge in specific clinical contexts or particular combinations of characteristics.

cs.AI↗

arXiv2025

Knowledge-Driven Feature Selection and Engineering for Genotype Data with Large Language Models

Predicting phenotypes with complex genetic bases based on a small, interpretable set of variant features remains a challenging task. Conventionally, data-driven approaches are utilized for this task, yet the high dimensional nature of genotype data makes the analysis and prediction difficult. Motivated by the extensive knowledge encoded in pre-trained LLMs and their success in processing complex biomedical concepts, we set to examine the ability of LLMs in feature selection and engineering for tabular genotype data, with a novel knowledge-driven framework. We develop FREEFORM, Free-flow Reasoning and Ensembling for Enhanced Feature Output and Robust Modeling, designed with chain-of-thought and ensembling principles, to select and engineer features with the intrinsic knowledge of LLMs. Evaluated on two distinct genotype-phenotype datasets, genetic ancestry and hereditary hearing loss, we find this framework outperforms several data-driven methods, particularly on low-shot regimes. FREEFORM is available as open-source framework at GitHub: https://github.com/PennShenLab/FREEFORM.

cs.LG↗

arXiv2025

Accessing the topological properties of human brain functional sub-circuits in Echo State Networks

Recent years have witnessed an emerging trend in neuromorphic computing that centers around the use of brain connectomics as a blueprint for artificial neural networks. Connectomics-based neuromorphic computing has primarily focused on embedding human brain large-scale structural connectomes (SCs), as estimated from diffusion Magnetic Resonance Imaging (dMRI) modality, to echo-state networks (ESNs). A critical step in ESN embedding requires pre-determined read-in and read-out layers constructed by the induced subgraphs of the embedded reservoir. As \textit{a priori} set of functional sub-circuits are derived from functional MRI (fMRI) modality, it is unknown, till this point, whether the embedding of fMRI-induced sub-circuits/networks onto SCs is well justified from the neuro-physiological perspective and ESN performance across a variety of tasks. This paper proposes a pipeline to implement and evaluate ESNs with various embedded topologies and processing/memorization tasks. To this end, we showed that different performance optimums highly depend on the neuro-physiological characteristics of these pre-determined fMRI-induced sub-circuits. In general, fMRI-induced sub-circuit-embedded ESN outperforms simple bipartite and various null models with feed-forward properties commonly seen in MLP for different tasks and reservoir criticality conditions. We provided a thorough analysis of the topological properties of pre-determined fMRI-induced sub-circuits and highlighted their graph-theoretical properties that play significant roles in determining ESN performance.

q-bio.NC↗

arXiv2025

Advances in RNA secondary structure prediction and RNA modifications: Methods, data, and applications

Due to the hierarchical organization of RNA structures and their pivotal roles in fulfilling RNA functions, the formation of RNA secondary structure critically influences many biological processes and has thus been a crucial research topic. This review sets out to explore the computational prediction of RNA secondary structure and its connections to RNA modifications, which have emerged as an active domain in recent years. We first examine the progression of RNA secondary structure prediction methodology, focusing on a set of representative works categorized into thermodynamic, comparative, machine learning, and hybrid approaches. Next, we survey the advances in RNA modifications and computational methods for identifying RNA modifications, focusing on the prominent modification types. Subsequently, we highlight the interplay between RNA modifications and secondary structures, emphasizing how modifications such as m6A dynamically affect RNA folding and vice versa. In addition, we also review relevant data sources and provide a discussion of current challenges and opportunities in the field. Ultimately, we hope our review will be able to serve as a cornerstone to aid in the development of innovative methods for this emerging topic and foster therapeutic applications in the future.

q-bio.BM↗

arXiv2024

DALK: Dynamic Co-Augmentation of LLMs and KG to answer Alzheimer's Disease Questions with Scientific Literature

Recent advancements in large language models (LLMs) have achieved promising performances across various applications. Nonetheless, the ongoing challenge of integrating long-tail knowledge continues to impede the seamless adoption of LLMs in specialized domains. In this work, we introduce DALK, a.k.a. Dynamic Co-Augmentation of LLMs and KG, to address this limitation and demonstrate its ability on studying Alzheimer's Disease (AD), a specialized sub-field in biomedicine and a global health priority. With a synergized framework of LLM and KG mutually enhancing each other, we first leverage LLM to construct an evolving AD-specific knowledge graph (KG) sourced from AD-related scientific literature, and then we utilize a coarse-to-fine sampling method with a novel self-aware knowledge retrieval approach to select appropriate knowledge from the KG to augment LLM inference capabilities. The experimental results, conducted on our constructed AD question answering (ADQA) benchmark, underscore the efficacy of DALK. Additionally, we perform a series of detailed analyses that can offer valuable insights and guidelines for the emerging topic of mutually enhancing KG and LLM. We will release the code and data at https://github.com/David-Li0406/DALK.

cs.CL↗

arXiv2024

Causality-based Subject and Task Fingerprints using fMRI Time-series Data

Recently, there has been a revived interest in system neuroscience causation models due to their unique capability to unravel complex relationships in multi-scale brain networks. In this paper, our goal is to verify the feasibility and effectiveness of using a causality-based approach for fMRI fingerprinting. Specifically, we propose an innovative method that utilizes the causal dynamics activities of the brain to identify the unique cognitive patterns of individuals (e.g., subject fingerprint) and fMRI tasks (e.g., task fingerprint). The key novelty of our approach stems from the development of a two-timescale linear state-space model to extract 'spatio-temporal' (aka causal) signatures from an individual's fMRI time series data. To the best of our knowledge, we pioneer and subsequently quantify, in this paper, the concept of 'causal fingerprint.' Our method is well-separated from other fingerprint studies as we quantify fingerprints from a cause-and-effect perspective, which are then incorporated with a modal decomposition and projection method to perform subject identification and a GNN-based (Graph Neural Network) model to perform task identification. Finally, we show that the experimental results and comparisons with non-causality-based methods demonstrate the effectiveness of the proposed methods. We visualize the obtained causal signatures and discuss their biological relevance in light of the existing understanding of brain functionalities. Collectively, our work paves the way for further studies on causal fingerprints with potential applications in both healthy controls and neurodegenerative diseases.

cs.LG↗

arXiv2024

A principled framework to assess the information-theoretic fitness of brain functional sub-circuits

In systems and network neuroscience, many common practices in brain connectomic analysis are often not properly scrutinized. One such practice is mapping a predetermined set of sub-circuits, like functional networks (FNs), onto subjects' functional connectomes (FCs) without adequately assessing the information-theoretic appropriateness of the partition. Another practice that goes unchallenged is thresholding weighted FCs to remove spurious connections without justifying the chosen threshold. This paper leverages recent theoretical advances in Stochastic Block Models (SBMs) to formally define and quantify the information-theoretic fitness (e.g., prominence) of a predetermined set of FNs when mapped to individual FCs under different fMRI task conditions. Our framework allows for evaluating any combination of FC granularity, FN partition, and thresholding strategy, thereby optimizing these choices to preserve important topological features of the human brain connectomes. By applying to the Human Connectome Project with Schaefer parcellations at multiple levels of granularity, the framework showed that the common thresholding value of 0.25 was indeed information-theoretically valid for group-average FCs despite its previous lack of justification. Our results pave the way for the proper use of FNs and thresholding methods and provide insights for future research in individualized parcellations.

q-bio.NC↗

arXiv2024

Volume-optimal persistence homological scaffolds of hemodynamic networks covary with MEG theta-alpha aperiodic dynamics

Higher-order properties of functional magnetic resonance imaging (fMRI) induced connectivity have been shown to unravel many exclusive topological and dynamical insights beyond pairwise interactions. Nonetheless, whether these fMRI-induced higher-order properties play a role in disentangling other neuroimaging modalities' insights remains largely unexplored and poorly understood. In this work, by analyzing fMRI data from the Human Connectome Project Young Adult dataset using persistent homology, we discovered that the volume-optimal persistence homological scaffolds of fMRI-based functional connectomes exhibited conservative topological reconfigurations from the resting state to attentional task-positive state. Specifically, while reflecting the extent to which each cortical region contributed to functional cycles following different cognitive demands, these reconfigurations were constrained such that the spatial distribution of cavities in the connectome is relatively conserved. Most importantly, such level of contributions covaried with powers of aperiodic activities mostly within the theta-alpha (4-12 Hz) band measured by magnetoencephalography (MEG). This comprehensive result suggests that fMRI-induced hemodynamics and MEG theta-alpha aperiodic activities are governed by the same functional constraints specific to each cortical morpho-structure. Methodologically, our work paves the way toward an innovative computing paradigm in multimodal neuroimaging topological learning.

q-bio.NC↗

arXiv2024

Dude: Dual Distribution-Aware Context Prompt Learning For Large Vision-Language Model

Prompt learning methods are gaining increasing attention due to their ability to customize large vision-language models to new domains using pre-trained contextual knowledge and minimal training data. However, existing works typically rely on optimizing unified prompt inputs, often struggling with fine-grained classification tasks due to insufficient discriminative attributes. To tackle this, we consider a new framework based on a dual context of both domain-shared and class-specific contexts, where the latter is generated by Large Language Models (LLMs) such as GPTs. Such dual prompt methods enhance the model's feature representation by joining implicit and explicit factors encoded in LLM knowledge. Moreover, we formulate the Unbalanced Optimal Transport (UOT) theory to quantify the relationships between constructed prompts and visual tokens. Through partial matching, UOT can properly align discrete sets of visual tokens and prompt embeddings under different mass distributions, which is particularly valuable for handling irrelevant or noisy elements, ensuring that the preservation of mass does not restrict transport solutions. Furthermore, UOT's characteristics integrate seamlessly with image augmentation, expanding the training sample pool while maintaining a reasonable distance between perturbed images and prompt inputs. Extensive experiments across few-shot classification and adapter settings substantiate the superiority of our model over current state-of-the-art baselines.

cs.CV↗

arXiv2024

Theorizing neuro-induced relationships between cognitive diversity, motivation, grit and academic performance in multidisciplinary engineering education context

Nowadays, engineers need to tackle many unprecedented challenges that are often complex, and, most importantly, cannot be exhaustively compartmentalized into a single engineering discipline. In other words, most engineering problems need to be solved from a multidisciplinary approach. However, conventional engineering programs usually adopt pedagogical approaches specifically tailored to traditional, niched engineering disciplines, which become increasingly deviated from the industry needs as those programs are typically designed and taught by instructors with highly specialized engineering training and credentials. To reduce the gap, more multidisciplinary engineering programs emerge by systematically stretching across all engineering fibers, and challenge the sub-optimal traditional pedagogy crowded in engineering classrooms. To further advance future-oriented pedagogy, in this work, we hypothesized neuro-induced linkages between how cognitively different learners are and how the linkages would affect learners in the knowledge acquisition process. We situate the neuro-induced linkages in the context of multidisciplinary engineering education and propose possible pedagogical approaches to actualize the implications of this conceptual framework. Our study, based on the innovative concept of brain fingerprint, would serve as a pioneer model to theorize key components of learner-centered multidisciplinary engineering pedagogy which centers on the key question: how do we motivate engineering students of different backgrounds from a neuro-inspired perspective?

cs.CY↗

arXiv2023

Tangent functional connectomes uncover more unique phenotypic traits

Functional connectomes (FCs) contain pairwise estimations of functional couplings based on pairs of brain regions activity. FCs are commonly represented as correlation matrices that are symmetric positive definite (SPD) lying on or inside the SPD manifold. Since the geometry on the SPD manifold is non-Euclidean, the inter-related entries of FCs undermine the use of Euclidean-based distances. By projecting FCs into a tangent space, we can obtain tangent functional connectomes (tangent-FCs). Tangent-FCs have shown a higher predictive power of behavior and cognition, but no studies have evaluated the effect of such projections with respect to fingerprinting. We hypothesize that tangent-FCs have a higher fingerprint than regular FCs. Fingerprinting was measured by identification rates (ID rates) on test-retest FCs as well as on monozygotic and dizygotic twins. Our results showed that identification rates are systematically higher when using tangent-FCs. Specifically, we found: (i) Riemann and log-Euclidean matrix references systematically led to higher ID rates. (ii) In tangent-FCs, Main-diagonal regularization prior to tangent space projection was critical for ID rate when using Euclidean distance, whereas barely affected ID rates when using correlation distance. (iii) ID rates were dependent on condition and fMRI scan length. (iv) Parcellation granularity was key for ID rates in FCs, as well as in tangent-FCs with fixed regularization, whereas optimal regularization of tangent-FCs mostly removed this effect. (v) Correlation distance in tangent-FCs outperformed any other configuration of distance on FCs or on tangent-FCs across the fingerprint gradient (here sampled by assessing test-retest, Monozygotic and Dizygotic twins). (vi)ID rates tended to be higher in task scans compared to resting-state scans when accounting for fMRI scan length.

q-bio.NC↗