SEARCH · PubMed Health

Results for “multimodal data”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources

PubMed2025

Building a Digital Health Research Platform to Enable Recruitment, Enrollment, Data Collection, and Follow-Up for a Highly Diverse Longitudinal US Cohort of 1 Million People in the All of Us Research Program: Design and Implementation Study.

BACKGROUND: Longitudinal cohort studies have traditionally relied on clinic-based recruitment models, which limit cohort diversity and the generalizability of research outcomes. Digital research platforms can be used to increase participant access, improve study engagement, streamline data collection, and increase data quality; however, the efficacy and sustainability of digitally enabled studies rely heavily on the design, implementation, and management of the digital platform being used. OBJECTIVE: We sought to design and build a secure, privacy-preserving, validated, participant-centric digital health research platform (DHRP) to recruit and enroll participants, collect multimodal data, and engage participants from diverse backgrounds in the National Institutes of Health's (NIH) All of Us Research Program (AOU). AOU is an ongoing national, multiyear study aimed to build a research cohort of 1 million participants that reflects the diversity of the United States, including minority, health-disparate, and other populations underrepresented in biomedical research (UBR). METHODS: We collaborated with community members, health care provider organizations (HPOs), and NIH leadership to design, build, and validate a secure, feature-rich digital platform to facilitate multisite, hybrid, and remote study participation and multimodal data collection in AOU. Participants were recruited by in-person, print, and online digital campaigns. Participants securely accessed the DHRP via web and mobile apps, either independently or with research staff support. The participant-facing tool facilitated electronic informed consent (eConsent), multisource data collection (eg, surveys, genomic results, wearables, and electronic health records [EHRs]), and ongoing participant engagement. We also built tools for research staff to conduct remote participant support, study workflow management, participant tracking, data analytics, data harmonization, and data management. RESULTS: We built a secure, participant-centric DHRP with engaging functionality used to recruit, engage, and collect data from 705,719 diverse participants throughout the United States. As of April 2024, 87% (n=613,976) of the participants enrolled via the platform were from UBR groups, including racial and ethnic minorities (n=282,429, 46%), rural dwelling individuals (n=49,118, 8%), those over the age of 65 years (n=190,333, 31%), and individuals with low socioeconomic status (n=122,795, 20%). CONCLUSIONS: We built a participant-centric digital platform with tools to enable engagement with individuals from different racial, ethnic, and socioeconomic backgrounds and other UBR groups. This DHRP demonstrated successful use among diverse participants. These findings could be used as best practices for the effective use of digital platforms to build and sustain cohorts of various study designs and increase engagement with diverse populations in health research.

Humans↗

PubMed1991

Integration of multimodality imaging data for radiotherapy treatment planning.

This paper describes computational techniques to permit the quantitative integration of magnetic resonance (MR), positron emission tomography (PET), and x-ray computed tomography (CT) imaging data sets. These methods are used to incorporate unique diagnostic information provided by PET and MR imaging into CT-based treatment planning for radiotherapy of intracranial tumors and vascular malformations. Integration of information from the different imaging modalities is treated as a two-step process. The first step is to determine the set of geometric parameters relating the coordinates of two imaging data sets. No universal method for determining these parameters is appropriate because of the diversity of contemporary imaging methods and data formats. Most situations can be handled by one of the four different techniques described. These four methods make use of specific geometric objects contained in the two data sets to determine the parameters. These objects are: (a) anatomical and/or fiducial points, (b) attached line markers, (c) anatomical surfaces, and (d) outlines of anatomical structures. The second step involves using the derived transformation to transfer outlines of treatment volumes and/or anatomical structures drawn on the images of one imaging study to the images of another study, usually the treatment planning CT. Solid modelling and image processing techniques have been adapted and developed further to accomplish this task. Clinical examples and phantom studies are presented which verify the different aspects of these techniques and demonstrate the accuracy with which they can be applied. Clinical use of these techniques for treatment planning has resulted in improvements in localization of treatment volumes and critical structures in the brain. These improvements have allowed greater sparing of normal tissues and more precise delivery of energy to the desired irradiation volume. It is believed that these improvements will have a positive impact on the outcome of radiation therapy.

Algorithms↗

PubMed2004

Combination of hyperbolic functions for multimodal biometrics data fusion.

In this paper, we treat the problem of combining fingerprint and speech biometric decisions as a classifier fusion problem. By exploiting the specialist capabilities of each classifier, a combined classifier may yield results which would not be possible in a single classifier. The Feedforward Neural Network provides a natural choice for such data fusion as it has been shown to be a universal approximator. However, the training process remains much to be a trial-and-error effort since no learning algorithm can guarantee convergence to optimal solution within finite iterations. In this work, we propose a network model to generate different combinations of the hyperbolic functions to achieve some approximation and classification properties. This is to circumvent the iterative training problem as seen in neural networks learning. In many decision data fusion applications, since individual classifiers or estimators to be combined would have attained a certain level of classification or approximation accuracy, this hyperbolic functions network can be used to combine these classifiers taking their decision outputs as the inputs to the network. The proposed hyperbolic functions network model is first applied to a function approximation problem to illustrate its approximation capability. This is followed by some case studies on pattern classification problems. The model is finally applied to combine the fingerprint and speaker verification decisions which show either better or comparable results with respect to several commonly used methods.

Algorithms↗

PubMed2026

AI-Supported, Integrative Prediction of Postoperative Delirium: Protocol for the CONFUSED Study.

BACKGROUND: Postoperative delirium (POD) is a frequent and serious complication in older surgical patients, characterized by acute cognitive dysfunction and fluctuating levels of consciousness. POD is associated with prolonged hospitalization, long-term cognitive decline, reduced quality of life, and increased mortality. Despite its clinical relevance, the underlying pathophysiological mechanisms remain poorly understood, and reliable biomarkers for early prediction and prevention are lacking. OBJECTIVE: The CONFUSED study aims to identify molecular and clinical predictors of POD by integrating clinical data with proteomic, transcriptomic, and epigenetic analyses. The primary objective is to develop predictive models for POD using multimodal data. Secondary objectives include the identification of delirium-associated genes, proteins, and epigenetic signatures, as well as the exploration of patient subgroups at increased risk for POD. METHODS: CONFUSED is a prospective observational cohort study conducted at a German university hospital. Adult patients undergoing major surgery under general anesthesia will be enrolled until 100 cases of POD have been observed, which is expected to require a total sample size of approximately 200 to 300 patients. Blood samples are collected at 4 predefined time points: before premedication, immediately after surgery, and on postoperative days 2 and 5. Samples undergo comprehensive proteomic profiling, transcriptomic analysis using RNA microarrays, DNA methylation analysis, and genotyping of selected polymorphisms. Clinical data, including demographics, comorbidities, perioperative variables, medications, and delirium assessments using the Confusion Assessment Method (CAM) and CAM for the intensive care unit, are systematically recorded. Statistical analyses include univariate and multivariate methods, as well as machine learning approaches such as random forests and support vector machines, to identify relevant biomarkers and develop predictive models. The study protocol follows STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) and TRIPOD (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis) guidelines and was approved by the responsible ethics committees. RESULTS: The study was registered in the German Clinical Trials Register (DRKS00033854) on March 18, 2024. Recruitment started in January 2024 and is ongoing at the time of manuscript submission. As of now, 135 patients have been enrolled. Sample collection and laboratory analyses are ongoing. Data analysis began in January 2026, with first results anticipated in July 2026. Final data lock is anticipated after the completion of recruitment. CONCLUSIONS: By integrating multimodal molecular data with clinical parameters and applying advanced machine learning techniques, the CONFUSED study aims to improve the prediction and understanding of POD. The results are expected to support the development of personalized preventive strategies and contribute to improved perioperative care for patients at risk of POD.

Humans↗

PubMed2026

transFusion: a novel comprehensive platform for integration analysis of single-cell and spatial transcriptomics.

MOTIVATION: Understanding spatial organization, intercellular interactions, and regulatory networks within the spatial context of tissues is crucial for uncovering complex biological processes and disease mechanisms. Spatial transcriptomics technologies have revolutionized this field by enabling the spatially resolved profiling of gene expression. 10× Visium has emerged as the predominant spatial technology, but its low resolution and the complexity of integrating multimodal datasets present significant analytical challenges, particularly for researchers with limited computational and statistical expertise. Current spatial transcriptomics analysis platforms generally fall short of effectively integrating multimodal data and maximizing the utility of spatial information-such as uncovering complex cellular spatial dependencies, multimodal gradient patterns, and spatial coexpression of ligand-receptor pairs and regulatory networks related to disease or biological states-thereby limiting their ability to provide comprehensive end-to-end analytical workflows when analyzing 10× Visium data. RESULTS: To address these limitations, we developed transFusion, a novel, advanced web-based platform specializing in the most comprehensive and effective integration analysis of scRNA-seq and 10× Visium spatial transcriptomics data. transFusion offers 12 key functions, from basic visualization to advanced analyses, including intercellular dependency analysis, ligand-receptor coexpression identification and visualization, and spatial multimodal gradient variation patterns. Two case studies were used to demonstrate transFusion's capabilities in exploring tissue architecture, intercellular communication, dependency networks, and multimodal gradient variation patterns with minimal computational skills and statistical expertise. transFusion provides a flexible and powerful framework for multimodal data integration analysis. AVAILABILITY AND IMPLEMENTATION: transFusion is freely available at https://github.com/WQLin8/transFusion.

Spatial Transcriptomics↗

PubMed2006

Multimodal visualization interface for data management, self-learning and data presentation.

A multimodal visualization software, called the Data Manager (DM), has been developed to increase interdisciplinary communication around the topic of visualization and modeling of various aspects of the human anatomy. Numerous tools used in Radiology are integrated in the interface that runs on standard personal computers. The available tools, combined to hierarchical data management and custom layouts, allow analyzing of medical imaging data using advanced features outside radiological premises (for example, for patient review, conference presentation or tutorial preparation). The system is free, and based on an open-source software development architecture, and therefore updates of the system for custom applications are possible.

Anatomy↗

PubMed2026

mmContext: an open framework for multimodal contrastive learning of omics and text data.

SUMMARY: Multimodal approaches are increasingly leveraged for integrating omics data with textual biological knowledge. Yet there is still no accessible, standardized framework that enables systematic comparison of omics representations with different text encoders within a unified workflow. We present mmContext, a lightweight and extensible multimodal embedding framework built on top of the open-source Sentence Transformers library. The software allows researchers to train or apply models that jointly embed omics and text data using any numeric representation stored in an AnnData.obsm layer and any text encoder available in Hugging Face. mmContext supports integration of diverse biological text sources and provides pipelines for training, evaluation, and data preparation. We train and evaluate models for a RNA-Seq and text integration task, and demonstrate their utility through zero-shot classification of cell types and diseases across four independent datasets. By releasing all models, datasets, and tutorials openly, mmContext enables reproducible and accessible multimodal learning for omics-text integration. AVAILABILITY AND IMPLEMENTATION: Pretrained checkpoints and full source code for our custom MMContextEncoder are available on Hugging Face huggingface.co/jo-mengr. The Python package github.com/mengerj/mmcontext provides the model implementation and training and evaluation scripts for custom training. The releases for the publication can be accessed via zenodo: adata_hf_datasets: doi.org/10.5281/zenodo.19185217 and mmContext: doi.org/10.5281/zenodo.19185493.

Computational Biology↗

PubMed2026

Breast Cancer Recurrence Status Assessment in 5 Years Using Multimodal Integrated Learning: A Feasibility Study.

Despite advances in breast cancer detection and treatment, recurrence after curative therapy continues to impact long-term survival and quality of life. Therefore, early identification of high-risk patients is crucial to guide personalized treatment and follow-up strategies. Although genomic assays provide valuable prognostic insights, their high cost and limited accessibility hinder widespread adoption in clinical practice. Recent machine learning or deep learning approaches leveraging clinical, imaging, or multimodal data have shown promise but do not reflect real-world clinical scenarios. This study proposes a deep learning-based multimodal framework for predicting 5-year breast cancer recurrence using routinely collected clinical data. The framework consists of three main components. First, we adopted automated tumor segmentation with MedSAM to extract the tumor region from ultrasound images. The radiomics features are extracted from those tumor regions. Second, report features are extracted using a Med-Contrastive Pre-trained Transformers (MedCPT)-based approach incorporating predefined, clinically informed queries. Third, a multimodal integration model jointly processes image, radiomics, clinical features, and report features through modality-specific branches. The image branch employs the Ultrasound Foundation Model (USFM) as the backbone, while structured tabular data is processed using the FT-Transformer architecture. The features of all branches are fused using a mixture-of-experts (MoE)-based classifier, and the entire model is trained using a progressive fusion training strategy. Experimental results confirm the feasibility of using ultrasound images with tumor mask integration for recurrence prediction and demonstrate the additive value of integrating multiple data modalities through the proposed multimodal integration model. The final model for recurrence prediction achieved an AUC of 0.7540, accuracy of 74.61%, sensitivity of 70.41%, and specificity of 76.44%. This feasibility study's findings underscore the potential of the proposed multimodal deep learning framework to provide accessible, accurate, and generalizable recurrence risk prediction using routinely available clinical data, potentially supporting more informed treatment decisions and personalized post-treatment monitoring in real-world clinical practice.

Breast cancer recurrence↗

PubMed2026

Next-Generation Disease Profiling by Integrating Histopathology with Spatial Multi-Omics Data.

The field of pathology has experienced several transformative changes in recent years with the advent of digital pathology and spatial multi-omics. These technologies have enhanced every aspect of pathology practice, from streamlining daily workflows to generating high-fidelity multi-omics data that provide pathologists with novel tools to refine disease profiling and clinical diagnosis. Each layer of multimodal data (genomic, metabolomic, proteomic, or transcriptomic) has uncovered a distinct facet of disease pathologies, and combined with machine learning/artificial intelligence-based data analysis and pattern recognition models, has provided holistic understanding of regulatory mechanisms underpinning them. However, high-dimensional data have far exceeded the volume, scale, and complexity of immunostaining methods implemented by pathologists and, thus, have generated significant challenges related to deconvolution, interpretation, and clinical translation. Furthermore, these multimodal studies have predominantly relied on computational methods to process data and extract disease-relevant insights, thus raising questions around relevance or role of a pathologist in this new era of multi-omics. This review will provide a perspective on the evolving fields of molecular histopathology and spatial -omics, leveraging them to approach disease profiling, and redefining the role of a pathologist during this process.

Humans↗

PubMed2004

A probabilistic functional atlas of the human subthalamic nucleus.

This paper introduces a method for generation and validation of a probabilistic functional brain atlas of subcortical structures from electrophysiological and neuroimaging data. The method contains three major steps: (1) acquisition of pre, intra, and postoperative multimodal data; (2) selection of an accurate data set for atlas generation; and (3) generation of the atlas from the selected data set. The method is applied to construct the probabilistic functional atlas of the human subthalamic nucleus (STN). The STN atlas has been built from data collected during surgical treatment of 184 patients with Parkinson's disease. It is based on preoperative X-ray ventriculography imaging, intraoperative electrophysiological measurements and X-ray imaging, and postoperative neurological assessment. The atlas features a high resolution of 0.25 mm(3) and a high accuracy of 0.25 mm. It is dynamic and can be rapidly recalculated for arbitrary resolution and extended by adding new patient data. The atlas can easily be reformatted and warped to match patient-specific data. Its applications include planning of subthalamic stimulations and neuroscience research to study functional properties of the STN. The presented method is general and can be applied for constructing human and animal probabilistic brain atlases.

Anatomy, Artistic↗

PubMed2026

Deep learning-based multimodal pathogenomics integration for precision cancer prognosis.

BACKGROUND: Recent studies have revealed valuable prognostic insights in haematoxylin and eosin (H&E)-stained histological sections and transcriptomic profiles, suggesting potential applications in machine learning. However, existing methods lack sufficient intra- and inter-modal interactions, and face challenges in clinical validation due to incomplete multimodal data. METHODS: We proposed PathoGems (PathoGenomics-based integrative survival prediction), a weakly-supervised, interpretable multimodal learning framework that integrates histology and genomic profiles for precise cancer prognosis prediction. To evaluate the robustness of PathoGems, we initially curated a dataset of 1965 cases across four cohorts from The Cancer Genome Atlas (TCGA), including breast, colorectal, glioblastoma, and esophageal cancers. For external validation, PathoGems was further evaluated on four independent cohorts, consisting of 76 breast cancer and 41 esophageal squamous cell carcinoma cases from Zhejiang Cancer Hospital, as well as 102 colorectal cancer and 58 glioblastoma cases from the Clinical Proteomic Tumor Analysis Consortium (CPTAC). RESULTS: PathoGems effectively stratified patients into favorable and unfavorable risk groups, revealing significant differences in histological patterns, genomic features, and overall survival (log-rank test, p < 0.05). Moreover, the model’s predictions are further supported by visualization and transcriptomic analysis, enhancing interpretability and reliability. CONCLUSIONS: By fusing histological and clinicogenomic multimodal models, PathoGems will provide a solid foundation for developing an innovative tool that aids clinicians in making informed decisions and selection personalized treatment strategies for cancer patients.

Humans↗

PubMed2006

Adaptive Markov modeling for mutual-information-based, unsupervised MRI brain-tissue classification.

This paper presents a novel method for brain-tissue classification in magnetic resonance (MR) images that relies on a very general, adaptive statistical model of image neighborhoods. The method models MR-tissue intensities as derived from stationary random fields. It models the associated Markov statistics nonparametrically via a data-driven strategy. This paper describes the essential theoretical aspects underpinning adaptive, nonparametric Markov modeling and the theory behind the consistency of such a model. This general formulation enables the method to easily adapt to various kinds of MR images and the associated acquisition artifacts. It implicitly accounts for the intensity nonuniformity and performs reasonably well on T1-weighted MR data without nonuniformity correction. The method minimizes an information-theoretic metric on the probability density functions associated with image neighborhoods to produce an optimal classification. It automatically tunes its important internal parameters based on the information content of the data. Combined with an atlas-based initialization, it is completely automatic. Experiments on real, simulated, and multimodal data demonstrate the advantages of the method over the current state-of-the-art.

Algorithms↗

PubMed2025

Q RadFusion: Hybrid Quantum Classical Radiogenomic Framework for Breast Cancer Diagnosis.

BACKGROUND AND PURPOSE: Breast cancer remains the most common cancer in women worldwide, with early and accurate diagnosis critical for patient survival. Radiogenomics integrates imaging phenotypes with genomic profiles, offering a pathway to precision diagnostics. However, existing classical machine learning models often struggle with the high dimensionality and heterogeneity of multimodal data, leading to issues in calibration and reproducibility. This study presents Q RadFusion, a hybrid quantum-classical framework designed to enhance breast cancer diagnosis by fusing mammography and genomics data. METHODS: Q RadFusion was implemented on two publicly available datasets: CBIS-DDSM (2,600 curated mammography cases, TCIA) and TCGA-BRCA (1,000 genomic profiles, GDC). Imaging preprocessing included bias-field correction, segmentation, and harmonization, while genomic data underwent normalization and imputation. Feature selection was performed using the Quantum Approximate Optimization Algorithm (QAOA), and features were mapped into a quantum Hilbert space using Variational Quantum Circuits (VQC). For multimodal fusion, ResNet encoded mammography features, and a Transformer encoded genomic features. Patient-level and site-held-out splits were used for evaluation. RESULTS: Q RadFusion achieved an AUC of 0.96 and accuracy of 94%, outperforming baselines including CNN-LSTM, ResNet + XGBoost, and multimodal Transformers. Ablation studies confirmed the contribution of quantum components, with optimal performance observed at circuit depth, qubits, and QAOA layers. The model also demonstrated improved calibration and ~ 80% fewer parameters compared to deep fusion networks. CONCLUSION: Q RadFusion demonstrates that hybrid quantum-classical radiogenomic integration can deliver accurate, reproducible, and clinically meaningful diagnostic support for breast cancer, with strong potential for future clinical translation.

Breast Cancer↗

PubMed2005

Tensorial extensions of independent component analysis for multisubject FMRI analysis.

We discuss model-free analysis of multisubject or multisession FMRI data by extending the single-session probabilistic independent component analysis model (PICA; Beckmann and Smith, 2004. IEEE Trans. on Medical Imaging, 23 (2) 137-152) to higher dimensions. This results in a three-way decomposition that represents the different signals and artefacts present in the data in terms of their temporal, spatial, and subject-dependent variations. The technique is derived from and compared with parallel factor analysis (PARAFAC; Harshman and Lundy, 1984. In Research methods for multimode data analysis, chapter 5, pages 122-215. Praeger, New York). Using simulated data as well as data from multisession and multisubject FMRI studies we demonstrate that the tensor PICA approach is able to efficiently and accurately extract signals of interest in the spatial, temporal, and subject/session domain. The final decompositions improve upon PARAFAC results in terms of greater accuracy, reduced interference between the different estimated sources (reduced cross-talk), robustness (against deviations of the data from modeling assumptions and against overfitting), and computational speed. On real FMRI 'activation' data, the tensor PICA approach is able to extract plausible activation maps, time courses, and session/subject modes as well as provide a rich description of additional processes of interest such as image artefacts or secondary activation patterns. The resulting data decomposition gives simple and useful representations of multisubject/multisession FMRI data that can aid the interpretation and optimization of group FMRI studies beyond what can be achieved using model-based analysis techniques.

Artifacts↗

PubMed2004

Hypotension-induced loss of intraoperative monitoring data during surgical correction of scheuermann kyphosis: a case report.

STUDY DESIGN: Presentation of a case report of Scheuermann kyphosis surgical correction. OBJECTIVE: To describe a scenario where both neurogenic mixed evoked potentials and somatosensory-evoked potentials were lost due solely to hypotension before any correction of a kyphotic spinal deformity was performed. SUMMARY OF BACKGROUND DATA: Multimodality intraoperative neurophysiologic monitoring of the spinal cord has become widely utilized during surgical correction of scoliotic and kyphotic deformities. Most spinal surgeries also benefit from a state of hypotension to minimize blood loss, but unchecked and persistent hypotension may lead to inadequate perfusion to the spinal cord, resulting in spinal cord dysfunction noted by diminution of neuromonitoring data. METHODS: An 18-year-old boy with a 95 degrees Scheuermann kyphosis underwent a posterior spinal fusion for correction of his deformity. Intraoperative neurophysiologic monitoring consisting of neurogenic mixed evoked potentials and somatosensory-evoked potentials were performed throughout surgery. RESULTS: After placement of segmental pedicle screw fixation points and multiple osteotomies, before any instrumented correction of the deformity, all lower extremity neuromonitoring data were acutely lost. The surgeon was immediately warned of the data loss, with the mean arterial pressure noted to be 50 mm Hg. The mean arterial pressure was raised with the use of epinephrine bolus and dopamine infusion. Subsequently, all lower extremity neuromonitoring data returned. A Stagnara wake-up test was performed, which the patient passed, and the surgical correction was performed with his pressure maintained on a dopamine infusion. He awakened without neurologic deficits and had an uneventful recovery. CONCLUSIONS: Although a state of mild hypotension may be beneficial to limit blood loss during spinal deformity corrective surgery, acute and/or prolonged hypotension may jeopardize spinal cord vascularity and should be avoided especially during surgical treatment of high-risk deformities such as kyphosis. Early warning by multimodality physiologic neuromonitoring appears to be a useful method to alert surgeons of the potentially devastating problem of hypotension-induced spinal cord dysfunction and allows immediate corrective actions.

Adolescent↗

PubMed1999

A fuzzy clustering based segmentation system as support to diagnosis in medical imaging.

In medical imaging uncertainty is widely present in data, because of the noise in acquisition and of the partial volume effects originating from the low resolution of sensors. In particular, borders between tissues are not exactly defined and memberships in the boundary regions are intrinsically fuzzy. Therefore, computer assisted unsupervised fuzzy clustering methods turn out to be particularly suitable for handling a decision making process concerning segmentation of multimodal medical images. By using the possibilistic c-means algorithm as a refinement of a neural network based clustering algorithm named capture effect neural network, we developed the possibilistic neuro fuzzy c-means algorithm (PNFCM). In this paper the PNFCM has been applied to two different multimodal data sets and the results have been compared to those obtained by using the classical fuzzy c-means algorithm. Furthermore, a discussion is presented about the role of fuzzy clustering as a support to diagnosis in medical imaging.

Algorithms↗

PubMed2026

Large language models in bioinformatics: a comprehensive survey.

The emergence of foundation models with trillion-level parameters has redefined the landscape of artificial intelligence. Various fields are developing their own large-scale models, which can solve many problems within the field and improve work efficiency. Biological large-scale models are a cross-disciplinary research field that combines mathematics, computer science, and biology, aiming to simulate and understand the structure, function, and dynamic changes of biological systems through the establishment of complex computational models. This field covers multiple levels such as biological pathways, population dynamics, protein folding, etc., providing us with tools for deep exploration of the mysteries of life and applications in medicine, ecology, and other fields. This article reviews the background and research status of biological large-scale models, and discusses future directions. Large language models (LLMs) and other large-scale foundation models have rapidly advanced in recent years, enabling powerful representation learning and generation across text, sequences, and multimodal data. In bioinformatics and biomedicine, these models are increasingly used to analyze genomic sequences, infer protein properties and structures, support drug discovery, and integrate heterogeneous biomedical evidence. This survey reviews the basic principles of LLMs and summarizes representative applications in (i) gene and genome sequence analysis, (ii) protein structure and function prediction, and (iii) drug design, including virtual screening and personalized medicine. We also discuss emerging multi-model modeling approaches, as well as key challenges such as data quality and privacy, interpretability, generalization to new organisms and tasks, and responsible deployment in health-related settings. Finally, we outline future directions for developing reliable, scalable, and explainable bioinformatics foundation models.

bioinformatics↗

PubMed2006

Simultaneous EEG-fMRI.

Acquisition of electroencephalogram (EEG) during functional magnetic resonance imaging (fMRI) provides an additional monitoring tool for the analysis of brain state fluctuations. The exploration of brain responses following inputs or in the context of state changes is crucial for a better understanding of the basic principles governing large-scale neuronal dynamics. State-of-the-art techniques allow EEG activity-from DC (direct current) up to high frequencies in the gamma range-to be acquired simultaneously with fMRI data. In the interleaved mode, spiking activities can also be assessed during concurrent fMRI. The utilization of fMRI evidence to better constrain solutions of the inverse problem of source localization of EEG activity is an exciting possibility. Nonetheless, this approach should be applied cautiously since the degree of overlap between underlying neuronal activity sources is variable and, for the most part, unknown. The ultimate goal is to make joint inferences about the activity, dynamics, and functions by exploiting complementary information from multimodal data sets.

Animals↗