Aug 2026· ACS Measurement Science Au· 0 citations· 30 references
TL;DR
A modality-aware comparative framework that enables direct, side-by-side evaluation of clinical-only, metabolomics-only, and combined modeling strategies within a unified pipeline, and reveals that the dominant contributing modality shifted by scenario, reflecting differences in disease context.
Abstract
Disease classification using clinical and metabolomics data increasingly relies on multimodal integration, yet the complementary and comparative contributions of these modalities remain poorly understood. Most existing frameworks prioritize predictive performance without systematically examining how data modalities and modeling paradigms influence classification outcomes. Consequently, the relative value of individual modalities versus their integration, particularly in terms of model behavior, robustness, and interpretability, remains poorly characterized across machine learning (ML) and deep learning (DL) approaches. Here, we present a modality-aware comparative framework that enables direct, side-by-side evaluation of clinical-only, metabolomics-only, and combined modeling strategies within a unified pipeline. Unlike existing tools designed primarily for multiomics integration, this framework explicitly assesses when and how each modality contributes value across diverse classification scenarios. It supports the efficient use of available data while enabling a systematic comparison of model performance, stability, and interpretability across ML and DL methods. Rather than introducing a new classifier model, this work delivers a unified benchmarking workflow for evidence-based decision-making of modeling strategies in small-sample clinical metabolomics settings. Applied to two glomerulonephritis cohorts representing clinically-driven and metabolomics-driven classification settings, the framework revealed that the dominant contributing modality shifted by scenario, reflecting differences in disease context. These scenarios reflect realistic situations where discriminative signals arise from clinical variables, metabolic alterations, or their combination, as well as more complex cases involving subclass discrimination with overlapping profiles. While several models achieved comparable predictive accuracy, they differed in feature ranking stability, sampling sensitivity, and tendency to overfitting. Overall, this framework facilitates transparent, evidence-based selection of modeling strategies and data modalities suited to data complexity, sample size, and research objectives. Source code is available at https://github.com/kwanjeeraw/MMFramework.
Effective glioblastoma care requires integrating multiparametric longitudinal MRI with histopathology, molecular profiling, and clinical records documenting surgery, radiotherapy, chemotherapy, and supportive treatments. In routine practice, however, these data streams are often evaluated separately rather than jointly, which can delay molecularly informed stratification, limit reproducibility across centers, and complicate interpretation of post-treatment imaging changes. Multimodal machine learning (MML) provides a framework for clinical decision support by integrating diverse patient data across the course of care, from symptom presentation through diagnosis to treatment decisions. By combining MRI, whole-slide pathology, molecular and methylation profiling, and treatment timelines derived from electronic health records, MML models can capture disease characteristics over time across biological scales through representation learning and multimodal fusion. Importantly, these approaches can incorporate uncertainty through model calibration and confidence-aware predictions. When rigorously developed and validated, MML models may generate clinically relevant outputs, including integrated diagnosis, molecular classification, individualized survival estimates, probabilistic discrimination between tumor progression and pseudo-progression, and stratification for clinical trial eligibility. In this Mini Review, we summarize recent advances and emerging translational evidence for clinically oriented MML in glioblastoma, with particular emphasis on MRI-centered systems that integrate imaging with pathology, selected molecular measurements, and longitudinal clinical context. We also outline key methodological and practical considerations, including dataset curation, leakage control, external validation, calibration, and post-deployment monitoring—required to support safe, robust, and generalizable implementation of MML approaches in neuro-oncology practice.
Amin Zadeh-Shirazi, Bryan W. Day, Hui K. Gan et al.· Frontiers in Oncology· 0 citations
MB-SupCon-cont improves prediction accuracy by incorporating a generalized contrastive loss function that defines similarity and dissimilarity for continuous responses using three distance-based weighting methods, and provides superior representation learning and improves data visualization in lower-dimensional spaces.
Sen Yang, Shidan Wang, Yiqing Wang et al.· Frontiers in microbiomes· 0 citations
Abstract Precision medicine is increasingly constrained not by a lack of molecular data but by the absence of frameworks that can translate multidimensional biological information into actionable clinical decisions. Multi‐omics‐driven precision medicine (MODPM) addresses this lack by integrating genomics, epigenomics, transcriptomics, proteomics, metabolomics, microbiome, and clinical context into a multiscale framework that links molecular mechanisms, tissue organization, and patient trajectories. In this review, we propose a conceptual framework for MODPM and examine how advances in multi‐omics technologies, artificial intelligence (AI), and foundation models are reshaping disease modeling, drug development, and precision intervention. We summarize the biological contributions of major omics layers and discuss how AI supports cross‐modal representation learning, contextual modeling, and perturbation‐aware prediction. We highlight drug development as a key translational application of MODPM and further discuss its clinical relevance across three major disease contexts: cancer, autoimmune diseases, and metabolic disorders, including cardiometabolic and renal–metabolic diseases. These examples illustrate how MODPM can support target discovery, disease endotyping, treatment response prediction, and clinical monitoring by analyzing shared mechanisms such as immune dysregulation, metabolic remodeling, chronic inflammation, tissue microenvironmental changes, and gene–environment interactions. Across these settings, MODPM enables finer molecular stratification, the identification of pathway‐dominant disease states, improved response prediction, and dynamic treatment monitoring. We also discuss key barriers to implementation, including data heterogeneity, limited cohort diversity, polygenic complexity, workflow constraints, cost, and ethical issues related to privacy, consent, and data ownership. Overall, the value of MODPM lies not in stacking additional data layers but in building a multiscale, continuously learnable framework to link biological heterogeneity to clinically interpretable and actionable decisions.
Machine learning is a ubiquitous part of the modern neuroimaging toolkit, particularly for research aimed towards precision medicine goals of improving individual-level diagnosis and treatment. However, the high dimensionality of neuroimaging data poses significant challenges for constructing interpretable predictive models. Several established methods, such as Connectome-based Predictive Modeling (CPM), Polyneuro Risk Scores (PNRS, inspired by Polygenic Risk Scores), and Polyconnectomic Scoring (PCS), offer an interpretable approach, which we term “Mass Univariate Aggregation” (MUA). MUA approaches evaluate each feature independently and then use a linear combination of weighted features to derive predictions, providing directly interpretable weights for individual features. Despite the existence and widespread usage of various MUA approaches in neuroimaging, tools for their application are still fragmented, and there does not yet exist an open-access unified tool for implementing, evaluating, or comparing these models within a standardized machine learning workflow. Here, we present a unified, flexible, and accessible configurable pipeline that can be used for implementing CPM, PNRS, PCS, and many new MUA configurations facilitated by user-specified parameters. Built in Python as an add-on for scikit-learn, our configurable pipeline enables researchers to leverage the functionality and standards provided by a widely used open-access machine learning tool. We validated the configurable pipeline by replicating the outcomes achieved by existing CPM and PNRS implementations, utilizing resting-state functional connectivity data from the Human Connectome Project to predict fluid intelligence (n = 1067). We further validated the pipeline’s PCS implementation, confirming PCS computation with external connectome summary statistics (CSS) matrices using the same data, and CSS derivation using simulated data. Although designed to fill a gap in neuroimaging, our open-source, configurable pipeline provides a standardized platform for applying the MUA methods to any machine learning setting that features high-dimensional data.
Machine learning (ML) integrated with metabolomics has emerged as a promising strategy to advance precision pharmacotherapy, enabling data-driven prediction of drug response. This review provides an overview of commonly applied ML methodologies in metabolomics-based pharmacological studies, including supervised models (Random Forest, Extreme Gradient Boosting, Support Vector Machine, Logistic Regression, K-Nearest Neighbors), unsupervised models (K-Means Clustering, Principal Component Analysis), and deep learning approaches. We summarize recent progress in the application of metabolomics-driven ML to personalized medication, with a focus on drug dosage optimization, therapeutic efficacy prediction, and adverse drug reaction assessment. Despite these advances, significant challenges remain, including limited explainability, insufficient prospective clinical validation, lack of standardization and reproducibility, and data dimensionality and quality issues. Addressing these issues will be essential for the clinical translation of ML-metabolomics integration. Looking ahead, continued methodological innovation, large-scale multi-center prospective validation, and integration with other omics platforms will be key to unlocking the full potential of metabolomics combined with ML in precision healthcare.
Pan Li, Jing Mao, Xiang-Lin Hu et al.· Metabolites· 0 citations
Accurate breast cancer prognosis remains a major challenge in precision oncology due to tumor heterogeneity and the complexity of integrating high-dimensional multi-omics data. Although multimodal learning approaches have improved predictive performance by combining clinical and molecular information, many existing methods rely on a single ensemble strategy that remains susceptible to prediction variance and limited robustness in high-dimensional, low-sample-size biomedical datasets. This study investigated whether integrating complementary ensemble strategies within a unified multimodal framework could improve the robustness and predictive performance of breast cancer prognosis. A heterogeneous multimodal ensemble framework was developed in which stacking was used to integrate complementary information from clinical, gene expression, and copy number variation (CNV) data through meta-learning, while bagging was incorporated to stabilize the meta-learning process via bootstrap aggregation. The outputs of the stacking and bagging branches were combined using weighted probability fusion. The framework was evaluated on the METABRIC breast cancer cohort and compared with unimodal models and a conventional stacking ensemble using an independent test set and stratified tenfold cross-validation. The proposed hybrid framework achieved a ROC-AUC of 0.936, outperforming unimodal clinical and molecular models (ROC-AUC = 0.8140.885) and the conventional stacking ensemble (ROC-AUC = 0.898). Stratified tenfold cross-validation further demonstrated consistent improvements in mean ROC-AUC, recall, F1-score, balanced accuracy, and Matthews correlation coefficient, indicating improved robustness and stable performance across the internal validation folds. On the independent test set, the hybrid framework reduced false-negative predictions and increased sensitivity relative to the stacking ensemble, demonstrating a more favorable balance between identifying high-risk patients and maintaining overall predictive performance. Rather than introducing a new ensemble algorithm, this study demonstrates that assigning complementary roles to stacking multimodal information integration and bagging for prediction stabilization provides an effective and robust framework for multi-omics breast cancer prognosis. The proposed hybrid strategy consistently improved predictive performance and robustness compared with conventional stacking while demonstrating stable performance across internal validation, supporting the use of complementary ensemble paradigms for multimodal prediction in precision oncology.
Reza Bozorgpour, Mohammadreza Soltany Sadrabadi· Clinical Cancer Bulletin· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
