Medical artificial intelligence (AI) systems frequently lack systematic domain expertise integration, potentially compromising diagnostic reliability. This study presents an ontology-based framework for bone disease diagnosis, developed in collaboration with Ho Chi Minh City Hospital for Traumatology and Orthopedics. The framework introduces three theoretical contributions: (1) a hierarchical neural network architecture guided by bone disease ontology for segmentation-classification tasks, incorporating Visual Language Models (VLMs) through prompts, (2) an ontology-enhanced Visual Question Answering (VQA) system for clinical reasoning, and (3) a multimodal deep learning model that integrates imaging, clinical, and laboratory data through ontological relationships. The methodology maintains clinical interpretability through systematic knowledge digitization, standardized medical terminology mapping, and modular architecture design. The framework demonstrates potential for extension beyond bone diseases through its standardized structure and reusable components. While theoretical foundations are established, experimental validation remains pending due to current dataset and computational resource limitations. Future work will focus on expanding the clinical dataset and conducting comprehensive system validation.
Gaining insights into the structural and functional mechanisms of the brain has been a longstanding focus in neuroscience research, particularly in the context of understanding and treating neuropsychiatric disorders such as Schizophrenia (SZ). Nevertheless, most of the traditional multimodal deep learning approaches fail to fully leverage the complementary characteristics of structural and functional connectomics data to enhance diagnostic performance. To address this issue, we proposed ConneX, a multimodal fusion method that integrates cross-attention mechanism and multilayer perceptron (MLP)-Mixer for refined feature fusion. Modality-specific backbone graph neural networks (GNNs) were firstly employed to obtain feature representation for each modality. A unified cross-modal attention network was then introduced to fuse these embeddings by capturing intra- and inter-modal interactions, while MLP-Mixer layers refined global and local features, leveraging higher-order dependencies for end-to-end classification with a multi-head joint loss. Extensive evaluations demonstrated improved performance on two distinct clinical datasets, highlighting the robustness of our proposed framework.
Emotion recognition is essential across numerous fields, including medical applications and brain-computer interface (BCI). Emotional responses include behavioral reactions, such as tone of voice and body movement, and changes in physiological signals, such as the electroencephalogram (EEG). The latter are involuntary, thus they provide a reliable input for identifying emotions, in contrast to the former which individuals can consciously control. These signals reveal true emotional states without intentional alteration, thus increasing the accuracy of emotion recognition models. However, multimodal deep learning methods from physiological signals have not been significantly investigated. In this paper, we introduce PHemoNet, a fully hypercomplex network for multimodal emotion recognition from physiological signals. In detail, the architecture comprises modality-specific encoders and a fusion module. Both encoders and fusion modules are defined in the hypercomplex domain through parameterized hypercomplex multiplications (PHMs) that can capture latent relations between the different dimensions of each modality and between the modalities themselves. The proposed method outperforms current state-of-the-art models on the MAHNOB-HCI dataset in classifying valence and arousal using electroencephalograms (EEGs) and peripheral physiological signals. The code for this work is available at https://github.com/ispamm/MHyEEG.
Effective dietary monitoring is critical for managing Type 2 diabetes, yet accurately estimating caloric intake remains a major challenge. While continuous glucose monitors (CGMs) offer valuable physiological data, they often fall short in capturing the full nutritional profile of meals due to inter-individual and meal-specific variability. In this work, we introduce a multimodal deep learning framework that jointly leverages CGM time-series data, Demographic/Microbiome, and pre-meal food images to enhance caloric estimation. Our model utilizes attention based encoding and a convolutional feature extraction for meal imagery, multi-layer perceptrons for CGM and Microbiome data followed by a late fusion strategy for joint reasoning. We evaluate our approach on a curated dataset of over 40 participants, incorporating synchronized CGM, Demographic and Microbiome data and meal photographs with standardized caloric labels. Our model achieves a Root Mean Squared Relative Error (RMSRE) of 0.2544, outperforming the baselines models by over 50%. These findings demonstrate the potential of multimodal sensing to improve automated dietary assessment tools for chronic disease management.
Stochastic Discount Factor (SDF) models provide a unified framework for asset pricing and risk assessment, yet traditional formulations struggle to incorporate unstructured textual information. We introduce NewsNet-SDF, a novel deep learning framework that seamlessly integrates pretrained language model embeddings with financial time series through adversarial networks. Our multimodal architecture processes financial news using GTE-multilingual models, extracts temporal patterns from macroeconomic data via LSTM networks, and normalizes firm characteristics, fusing these heterogeneous information sources through an innovative adversarial training mechanism. Our dataset encompasses approximately 2.5 million news articles and 10,000 unique securities, addressing the computational challenges of processing and aligning text data with financial time series. Empirical evaluations on U.S. equity data (1980-2022) demonstrate NewsNet-SDF substantially outperforms alternatives with a Sharpe ratio of 2.80. The model shows a 471% improvement over CAPM, over 200% improvement versus traditional SDF implementations, and a 74% reduction in pricing errors compared to the Fama-French five-factor model. In comprehensive comparisons, our deep learning approach consistently outperforms traditional, modern, and other neural asset pricing models across all key metrics. Ablation studies confirm that text embeddings contribute significantly more to model performance than macroeconomic features, with news-derived principal components ranking among the most influential determinants of SDF dynamics. These results validate the effectiveness of our multimodal deep learning approach in integrating unstructured text with traditional financial data for more accurate asset pricing, providing new insights for digital intelligent decision-making in financial technology.
Academic misconduct detection in biomedical research remains challenging due to algorithmic narrowness in existing methods and fragmented analytical pipelines. We present BMMDetect, a multimodal deep learning framework that integrates journal metadata (SJR, institutional data), semantic embeddings (PubMedBERT), and GPT-4o-mined textual attributes (methodological statistics, data anomalies) for holistic manuscript evaluation. Key innovations include: (1) multimodal fusion of domain-specific features to reduce detection bias; (2) quantitative evaluation of feature importance, identifying journal authority metrics (e.g., SJR-index) and textual anomalies (e.g., statistical outliers) as dominant predictors; and (3) the BioMCD dataset, a large-scale benchmark with 13,160 retracted articles and 53,411 controls. BMMDetect achieves 74.33% AUC, outperforming single-modality baselines by 8.6%, and demonstrates transferability across biomedical subfields. This work advances scalable, interpretable tools for safeguarding research integrity.
Integrated sensing and communication (ISAC) systems operating at terahertz (THz) bands are envisioned to enable both ultra-high data-rate communication and precise environmental awareness for next-generation wireless networks. However, the narrow width of THz beams makes them prone to misalignment and necessitates frequent beam prediction in dynamic environments. Multimodal sensing, which integrates complementary modalities such as camera images, positional data, and radar measurements, has recently emerged as a promising solution for proactive beam prediction. Nevertheless, existing multimodal approaches typically employ static fusion architectures that cannot adjust to varying modality reliability and contributions, thereby degrading predictive performance and robustness. To address this challenge, we propose a novel and efficient multimodal mixture-of-experts (MoE) deep learning framework for proactive beam prediction in THz ISAC systems. The proposed multimodal MoE framework employs multiple modality-specific expert networks to extract representative features from individual sensing modalities, and dynamically fuses them using adaptive weights generated by a gating network according to the instantaneous reliability of each modality. Simulation results in realistic vehicle-to-infrastructure (V2I) scenarios demonstrate that the proposed MoE framework outperforms traditional static fusion methods and unimodal baselines in terms of prediction accuracy and adaptability, highlighting its potential in practical THz ISAC systems with ultra-massive multiple-input multiple-output (MIMO).
Shape measures have emerged as promising descriptors of white matter tractography, offering complementary insights into anatomical variability and associations with cognitive and clinical phenotypes. However, conventional methods for computing shape measures are computationally expensive and time-consuming for large-scale datasets due to reliance on voxel-based representations. We propose Tract2Shape, a novel multimodal deep learning framework that leverages geometric (point cloud) and scalar (tabular) features to predict ten white matter tractography shape measures. To enhance model efficiency, we utilize a dimensionality reduction algorithm for the model to predict five primary shape components. The model is trained and evaluated on two independently acquired datasets, the HCP-YA dataset, and the PPMI dataset. We evaluate the performance of Tract2Shape by training and testing it on the HCP-YA dataset and comparing the results with state-of-the-art models. To further assess its robustness and generalization ability, we also test Tract2Shape on the unseen PPMI dataset. Tract2Shape outperforms SOTA deep learning models across all ten shape measures, achieving the highest average Pearson's r and the lowest nMSE on the HCP-YA dataset. The ablation study shows that both multimodal input and PCA contribute to performance gains. On the unseen testing PPMI dataset, Tract2Shape maintains a high Pearson's r and low nMSE, demonstrating strong generalizability in cross-dataset evaluation. Tract2Shape enables fast, accurate, and generalizable prediction of white matter shape measures from tractography data, supporting scalable analysis across datasets. This framework lays a promising foundation for future large-scale white matter shape analysis.
Stroke is a major public health problem, affecting millions worldwide. Deep learning has recently demonstrated promise for enhancing the diagnosis and risk prediction of stroke. However, existing methods rely on costly medical imaging modalities, such as computed tomography. Recent studies suggest that retinal imaging could offer a cost-effective alternative for cerebrovascular health assessment due to the shared clinical pathways between the retina and the brain. Hence, this study explores the impact of leveraging retinal images and clinical data for stroke detection and risk prediction. We propose a multimodal deep neural network that processes Optical Coherence Tomography (OCT) and infrared reflectance retinal scans, combined with clinical data, such as demographics, vital signs, and diagnosis codes. We pretrained our model using a self-supervised learning framework using a real-world dataset consisting of $37$ k scans, and then fine-tuned and evaluated the model using a smaller labeled subset. Our empirical findings establish the predictive ability of the considered modalities in detecting lasting effects in the retina associated with acute stroke and forecasting future risk within a specific time horizon. The experimental results demonstrate the effectiveness of our proposed framework by achieving $5$\% AUROC improvement as compared to the unimodal image-only baseline, and $8$\% improvement compared to an existing state-of-the-art foundation model. In conclusion, our study highlights the potential of retinal imaging in identifying high-risk patients and improving long-term outcomes.
This study proposes a novel approach combining Multimodal Deep Learning with intrinsic eXplainable Artificial Intelligence techniques to predict pathological response in non-small cell lung cancer patients undergoing neoadjuvant therapy. Due to the limitations of existing radiomics and unimodal deep learning approaches, we introduce an intermediate fusion strategy that integrates imaging and clinical data, enabling efficient interaction between data modalities. The proposed Multimodal Doctor-in-the-Loop method further enhances clinical relevance by embedding clinicians' domain knowledge directly into the training process, guiding the model's focus gradually from broader lung regions to specific lesions. Results demonstrate improved predictive accuracy and explainability, providing insights into optimal data integration strategies for clinical applications.
Multimodal deep learning harnesses diverse imaging modalities, such as MRI sequences, to enhance diagnostic accuracy in medical imaging. A key challenge is determining the optimal timing for integrating these modalities-specifically, identifying the network layers where fusion modules should be inserted. Current approaches often rely on manual tuning or exhaustive search, which are computationally expensive without any guarantee of converging to optimal results. We propose a sequential forward search algorithm that incrementally activates and evaluates candidate fusion modules at different layers of a multimodal network. At each step, the algorithm retrains from previously learned weights and compares validation loss to identify the best-performing configuration. This process systematically reduces the search space, enabling efficient identification of the optimal fusion timing without exhaustively testing all possible module placements. The approach is validated on two multimodal MRI datasets, each addressing different classification tasks. Our algorithm consistently identified configurations that outperformed unimodal baselines, late fusion, and a brute-force ensemble of all potential fusion placements. These architectures demonstrated superior accuracy, F-score, and specificity while maintaining competitive or improved AUC values. Furthermore, the sequential nature of the search significantly reduced computational overhead, making the optimization process more practical. By systematically determining the optimal timing to fuse imaging modalities, our method advances multimodal deep learning for medical imaging. It provides an efficient and robust framework for fusion optimization, paving the way for improved clinical decision-making and more adaptable, scalable architectures in medical AI applications.
Effective recognition of acute and difficult-to-heal wounds is a necessary step in wound diagnosis. An efficient classification model can help wound specialists classify wound types with less financial and time costs and also help in deciding on the optimal treatment method. Traditional machine learning models suffer from feature selection and are usually cumbersome models for accurate recognition. Recently, deep learning (DL) has emerged as a powerful tool in wound diagnosis. Although DL seems promising for wound type recognition, there is still a large scope for improving the efficiency and accuracy of the model. In this study, a DL-based multimodal classifier was developed using wound images and their corresponding locations to classify them into multiple classes, including diabetic, pressure, surgical, and venous ulcers. A body map was also created to provide location data, which can help wound specialists label wound locations more effectively. The model uses a Vision Transformer to extract hierarchical features from input images, a Discrete Wavelet Transform (DWT) layer to capture low and high frequency components, and a Transformer to extract spatial features. The number of neurons and weight vector optimization were performed using three swarm-based optimization techniques (Monster Gorilla Toner (MGTO), Improved Gray Wolf Optimization (IGWO), and Fox Optimization Algorithm). The evaluation results show that weight vector optimization using optimization algorithms can increase diagnostic accuracy and make it a very effective approach for wound detection. In the classification using the original body map, the proposed model was able to achieve an accuracy of 0.8123 using image data and an accuracy of 0.8007 using a combination of image data and wound location. Also, the accuracy of the model in combination with the optimization models varied from 0.7801 to 0.8342.
Inspired by human visual attention, deep neural networks have widely adopted attention mechanisms to learn locally discriminative attributes for challenging visual classification tasks. However, existing approaches primarily emphasize the representation of such features while neglecting their precise localization, which often leads to misclassification caused by shortcut biases. This limitation becomes even more pronounced when models are evaluated on transfer or out-of-distribution datasets. In contrast, humans are capable of leveraging prior object knowledge to quickly localize and compare fine-grained attributes, a capability that is especially crucial in complex and high-variance classification scenarios. Motivated by this, we introduce Gaze-CIFAR-10, a human gaze time-series dataset, along with a dual-sequence gaze encoder that models the precise sequential localization of human attention on distinct local attributes. In parallel, a Vision Transformer (ViT) is employed to learn the sequential representation of image content. Through cross-modal fusion, our framework integrates human gaze priors with machine-derived visual sequences, effectively correcting inaccurate localization in image feature representations. Extensive qualitative and quantitative experiments demonstrate that gaze-guided cognitive cues significantly enhance classification accuracy.
Objective: Neonates are highly susceptible to seizures, which can have severe long-term consequences if undetected and left untreated. Early detection is crucial and typically requires continuous electroencephalography (EEG) monitoring in a hospital setting, which is costly, inconvenient, and requires specialized experts for diagnosis. In this work, we propose a new low-cost active dry-contact electrode-based adjustable EEG headset, a new explainable deep learning model to detect neonatal seizures, and an advanced signal processing algorithm to remove artifacts to address the key aspects that lead to the underdiagnosis of neonatal seizures. Methods: EEG signals are acquired through active electrodes and processed using a custom-designed analog front end (AFE) that filters and digitizes the captured EEG signals. The adjustable headset is designed using three-dimensional (3D) printing and laser cutting to fit a wide range of head sizes. A deep learning model is developed to classify seizure and non-seizure epochs in real-time. Furthermore, a separate multimodal deep learning model is designed to remove noise artifacts. The device is tested on a pediatric patient with absence seizures in a hospital setting. Simultaneous recordings are captured using both the custom device and the commercial wet electrode device available in the hospital for comparison. Results: The signals obtained using our custom design and a commercial device show a high correlation (>0.8). Further analysis using signal-to-noise ratio values shows that our device can mitigate noise similar to the commercial device. The proposed deep learning model has improvements in accuracy and recall by 2.76% and 16.33%, respectively, compared to the state-of-the-art. Furthermore, the developed artifact removal algorithm can identify and remove artifacts while keeping seizure patterns intact.
Molecular subtyping of breast cancer is crucial for personalized treatment and prognosis. Traditional classification approaches rely on either histopathological images or gene expression profiling, limiting their predictive power. In this study, we propose a deep multimodal learning framework that integrates histopathological images and gene expression data to classify breast cancer into BRCA.Luminal and BRCA.Basal / Her2 subtypes. Our approach employs a ResNet-50 model for image feature extraction and fully connected layers for gene expression processing, with a cross-attention fusion mechanism to enhance modality interaction. We conduct extensive experiments using five-fold cross-validation, demonstrating that our multimodal integration outperforms unimodal approaches in terms of classification accuracy, precision-recall AUC, and F1-score. Our findings highlight the potential of deep learning for robust and interpretable breast cancer subtype classification, paving the way for improved clinical decision-making.
Acute Coronary Syndromes (ACS), including ST-segment elevation myocardial infarctions (STEMI) and non-ST-segment elevation myocardial infarctions (NSTEMI), remain a leading cause of mortality worldwide. Traditional cardiovascular risk scores rely primarily on clinical data, often overlooking environmental influences like air pollution that significantly impact heart health. Moreover, integrating complex time-series environmental data with clinical records is challenging. We introduce TabulaTime, a multimodal deep learning framework that enhances ACS risk prediction by combining clinical risk factors with air pollution data. TabulaTime features three key innovations: First, it integrates time-series air pollution data with clinical tabular data to improve prediction accuracy. Second, its PatchRWKV module automatically extracts complex temporal patterns, overcoming limitations of traditional feature engineering while maintaining linear computational complexity. Third, attention mechanisms enhance interpretability by revealing interactions between clinical and environmental factors. Experimental results show that TabulaTime improves prediction accuracy by over 20% compared to conventional models such as CatBoost, Random Forest, and LightGBM, with air pollution data alone contributing over a 10% improvement. Feature importance analysis identifies critical predictors including previous angina, systolic blood pressure, PM10, and NO2. Overall, TabulaTime bridges clinical and environmental insights, supporting personalized prevention strategies and informing public health policies to mitigate ACS risk.
Multimodal deep learning systems are deployed in dynamic scenarios due to the robustness afforded by multiple sensing modalities. Nevertheless, they struggle with varying compute resource availability (due to multi-tenancy, device heterogeneity, etc.) and fluctuating quality of inputs (from sensor feed corruption, environmental noise, etc.). Current multimodal systems employ static resource provisioning and cannot easily adapt when compute resources change over time. Additionally, their reliance on processing sensor data with fixed feature extractors is ill-equipped to handle variations in modality quality. Consequently, uninformative modalities, such as those with high noise, needlessly consume resources better allocated towards other modalities. We propose ADMN, a layer-wise Adaptive Depth Multimodal Network capable of tackling both challenges - it adjusts the total number of active layers across all modalities to meet compute resource constraints, and continually reallocates layers across input modalities according to their modality quality. Our evaluations showcase ADMN can match the accuracy of state-of-the-art networks while reducing up to 75% of their floating-point operations.
Data-driven susceptibility mapping of natural hazards has harnessed the advances in classification methods used on heterogeneous sources represented as raster images. Susceptibility mapping is an important step towards risk assessment for any natural hazard. Increasingly, multiple hazards co-occur spatially, temporally, or both, which calls for an in-depth study on multi-hazard susceptibility mapping. In recent years, single-hazard susceptibility mapping algorithms have become well-established and have been extended to multi-hazard susceptibility mapping. Deep learning is also emerging as a promising method for single-hazard susceptibility mapping. Here, we discuss the evolution of methods for a single hazard, their extensions to multi-hazard maps as a late fusion of decisions, and the use of deep learning methods in susceptibility mapping. We finally propose a vision for adapting data fusion strategies in multimodal deep learning to multi-hazard susceptibility mapping. From the background study of susceptibility methods, we demonstrate that deep learning models are promising, untapped methods for multi-hazard susceptibility mapping. Data fusion strategies provide a larger space of deep learning models applicable to multi-hazard susceptibility mapping.
Neural networks are one tool for approximating non-linear differential equations used in scientific computing tasks such as surrogate modeling, real-time predictions, and optimal control. PDE foundation models utilize neural networks to train approximations to multiple differential equations simultaneously and are thus a general purpose solver that can be adapted to downstream tasks. Current PDE foundation models focus on either learning general solution operators and/or the governing system of equations, and thus only handle numerical or symbolic modalities. However, real-world applications may require more flexible data modalities, e.g. text analysis or descriptive outputs. To address this gap, we propose a novel multimodal deep learning approach that leverages a transformer-based architecture to approximate solution operators for a wide variety of ODEs and PDEs. Our method integrates numerical inputs, such as equation parameters and initial conditions, with text descriptions of physical processes or system dynamics. This enables our model to handle settings where symbolic representations may be incomplete or unavailable. In addition to providing accurate numerical predictions, our approach generates interpretable scientific text descriptions, offering deeper insights into the underlying dynamics and solution properties. The numerical experiments show that our model provides accurate solutions for in-distribution data (with average relative error less than 3.3%) and out-of-distribution data (average relative error less than 7.8%) together with precise text descriptions (with correct descriptions generated 100% of times). In certain tests, the model is also shown to be capable of extrapolating solutions in time.
Individual's general well-being is greatly impacted by mental health conditions including depression and Post-Traumatic Stress Disorder (PTSD), underscoring the importance of early detection and precise diagnosis in order to facilitate prompt clinical intervention. An advanced multimodal deep learning system for the automated classification of PTSD and depression is presented in this paper. Utilizing textual and audio data from clinical interview datasets, the method combines features taken from both modalities by combining the architectures of LSTM (Long Short Term Memory) and BiLSTM (Bidirectional Long Short-Term Memory).Although text features focus on speech's semantic and grammatical components; audio features capture vocal traits including rhythm, tone, and pitch. This combination of modalities enhances the model's capacity to identify minute patterns connected to mental health conditions. Using test datasets, the proposed method achieves classification accuracies of 92% for depression and 93% for PTSD, outperforming traditional unimodal approaches and demonstrating its accuracy and robustness.