Promoting cross-modal representations to improve multimodal foundation models for physiological signals

• URL: http://arxiv.org/abs/2410.16424v1
• Date: Mon, 21 Oct 2024 18:47:36 GMT
• Title: Promoting cross-modal representations to improve multimodal foundation models for physiological signals
• Authors: Ching Fang, Christopher Sandino, Behrooz Mahasseni, Juri Minxha, Hadi Pouransari, Erdrin Azemi, Ali Moin, Ellen Zippi,
• Abstract summary: We use a masked autoencoding objective to pretrain a multimodal model. We show that the model learns representations that can be linearly probed for a diverse set of downstream tasks. We argue that explicit methods for inducing cross-modality may enhance multimodal pretraining strategies.
• Score: 3.630706646160043
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Many healthcare applications are inherently multimodal, involving several physiological signals. As sensors for these signals become more common, improving machine learning methods for multimodal healthcare data is crucial. Pretraining foundation models is a promising avenue for success. However, methods for developing foundation models in healthcare are still in early exploration and it is unclear which pretraining strategies are most effective given the diversity of physiological signals. This is partly due to challenges in multimodal health data: obtaining data across many patients is difficult and costly, there is a lot of inter-subject variability, and modalities are often heterogeneously informative across downstream tasks. Here, we explore these challenges in the PhysioNet 2018 dataset. We use a masked autoencoding objective to pretrain a multimodal model. We show that the model learns representations that can be linearly probed for a diverse set of downstream tasks. We hypothesize that cross-modal reconstruction objectives are important for successful multimodal training, as they encourage the model to integrate information across modalities. We demonstrate that modality dropout in the input space improves performance across downstream tasks. We also find that late-fusion models pretrained with contrastive learning objectives are less effective across multiple tasks. Finally, we analyze the model's representations, showing that attention weights become more cross-modal and temporally aligned with our pretraining strategy. The learned embeddings also become more distributed in terms of the modalities encoded by each unit. Overall, our work demonstrates the utility of multimodal foundation models with health data, even across diverse physiological data sources. We further argue that explicit methods for inducing cross-modality may enhance multimodal pretraining strategies.

Related papers

• Multimodal Masked Autoencoder Pre-training for 3D MRI-Based Brain Tumor Analysis with Missing Modalities [0.0]
  BM-MAE is a masked image modeling pre-training strategy tailored for multimodal MRI data. It seamlessly adapts to any combination of available modalities, extracting rich representations that capture both intra- and inter-modal information. It can quickly and efficiently reconstruct missing modalities, highlighting its practical value.
  arXiv  Detail & Related papers  (2025-05-01T14:51:30Z)
• Towards Robust Multimodal Physiological Foundation Models: Handling Arbitrary Missing Modalities [9.785262633953794]
  Physio Omni is a foundation model for multimodal physiological signal analysis. It trains a decoupled multimodal tokenizer, enabling masked signal pre-training. It achieves state-of-the-art performance while maintaining strong robustness to missing modalities.
  arXiv  Detail & Related papers  (2025-04-28T09:00:04Z)
• Continual Multimodal Contrastive Learning [70.60542106731813]
  Multimodal contrastive learning (MCL) advances in aligning different modalities and generating multimodal representations in a joint space. However, a critical yet often overlooked challenge remains: multimodal data is rarely collected in a single process, and training from scratch is computationally expensive. In this paper, we formulate CMCL through two specialized principles of stability and plasticity. We theoretically derive a novel optimization-based method, which projects updated gradients from dual sides onto subspaces where any gradient is prevented from interfering with the previously learned knowledge.
  arXiv  Detail & Related papers  (2025-03-19T07:57:08Z)
• What are You Looking at? Modality Contribution in Multimodal Medical Deep Learning Methods [0.13194391758295113]
  We present a method that measures the importance of each modality in a dataset for the model to fulfill its task. We found that some networks have modality preferences that tend to unimodal collapses, while some datasets are imbalanced from the ground up. With our method we make a crucial contribution to the field of interpretability in deep learning based multimodal research.
  arXiv  Detail & Related papers  (2025-02-28T12:39:39Z)
• MIND: Modality-Informed Knowledge Distillation Framework for Multimodal Clinical Prediction Tasks [50.98856172702256]
  We propose the Modality-INformed knowledge Distillation (MIND) framework, a multimodal model compression approach. MIND transfers knowledge from ensembles of pre-trained deep neural networks of varying sizes into a smaller multimodal student. We evaluate MIND on binary and multilabel clinical prediction tasks using time series data and chest X-ray images.
  arXiv  Detail & Related papers  (2025-02-03T08:50:00Z)
• Continually Evolved Multimodal Foundation Models for Cancer Prognosis [50.43145292874533]
  Cancer prognosis is a critical task that involves predicting patient outcomes and survival rates. Previous studies have integrated diverse data modalities, such as clinical notes, medical images, and genomic data, leveraging their complementary information. Existing approaches face two major limitations. First, they struggle to incorporate newly arrived data with varying distributions into training, such as patient records from different hospitals. Second, most multimodal integration methods rely on simplistic concatenation or task-specific pipelines, which fail to capture the complex interdependencies across modalities.
  arXiv  Detail & Related papers  (2025-01-30T06:49:57Z)
• Towards Foundation Models for Critical Care Time Series [38.09906416210531]
  We introduce a harmonized dataset for sequence modeling and transfer learning research, representing the first large-scale collection to include core treatment variables. Future plans involve expanding this dataset to support further advancements in transfer learning and the development of scalable, generalizable models for critical healthcare applications.
  arXiv  Detail & Related papers  (2024-11-25T12:49:55Z)
• Cross-Modal Few-Shot Learning: a Generative Transfer Learning Framework [58.362064122489166]
  This paper introduces the Cross-modal Few-Shot Learning task, which aims to recognize instances from multiple modalities when only a few labeled examples are available. We propose a Generative Transfer Learning framework consisting of two stages: the first involves training on abundant unimodal data, and the second focuses on transfer learning to adapt to novel data. Our finds demonstrate that GTL has superior performance compared to state-of-the-art methods across four distinct multi-modal datasets.
  arXiv  Detail & Related papers  (2024-10-14T16:09:38Z)
• HyperMM : Robust Multimodal Learning with Varying-sized Inputs [4.377889826841039]
  HyperMM is an end-to-end framework designed for learning with varying-sized inputs. We introduce a novel strategy for training a universal feature extractor using a conditional hypernetwork. We experimentally demonstrate the advantages of our method in two tasks: Alzheimer's disease detection and breast cancer classification.
  arXiv  Detail & Related papers  (2024-07-30T12:13:18Z)
• MMA-DFER: MultiModal Adaptation of unimodal models for Dynamic Facial Expression Recognition in-the-wild [81.32127423981426]
  Multimodal emotion recognition based on audio and video data is important for real-world applications. Recent methods have focused on exploiting advances of self-supervised learning (SSL) for pre-training of strong multimodal encoders. We propose a different perspective on the problem and investigate the advancement of multimodal DFER performance by adapting SSL-pre-trained disjoint unimodal encoders.
  arXiv  Detail & Related papers  (2024-04-13T13:39:26Z)
• HEALNet: Multimodal Fusion for Heterogeneous Biomedical Data [10.774128925670183]
  This paper presents the Hybrid Early-fusion Attention Learning Network (HEALNet), a flexible multimodal fusion architecture. We conduct multimodal survival analysis on Whole Slide Images and Multi-omic data on four cancer datasets from The Cancer Genome Atlas (TCGA) HEALNet achieves state-of-the-art performance compared to other end-to-end trained fusion models.
  arXiv  Detail & Related papers  (2023-11-15T17:06:26Z)
• Dynamic Multimodal Information Bottleneck for Multimodality Classification [26.65073424377933]
  We propose a dynamic multimodal information bottleneck framework for attaining a robust fused feature representation. Specifically, our information bottleneck module serves to filter out the task-irrelevant information and noises in the fused feature. Our method surpasses the state-of-the-art and is significantly more robust, being the only method to remain performance when large-scale noisy channels exist.
  arXiv  Detail & Related papers  (2023-11-02T08:34:08Z)
• Learning Unseen Modality Interaction [54.23533023883659]
  Multimodal learning assumes all modality combinations of interest are available during training to learn cross-modal correspondences. We pose the problem of unseen modality interaction and introduce a first solution. It exploits a module that projects the multidimensional features of different modalities into a common space with rich information preserved.
  arXiv  Detail & Related papers  (2023-06-22T10:53:10Z)
• Learnable Weight Initialization for Volumetric Medical Image Segmentation [66.3030435676252]
  We propose a learnable weight-based hybrid medical image segmentation approach. Our approach is easy to integrate into any hybrid model and requires no external training data. Experiments on multi-organ and lung cancer segmentation tasks demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2023-06-15T17:55:05Z)
• Understanding the Tricks of Deep Learning in Medical Image Segmentation: Challenges and Future Directions [66.40971096248946]
  In this paper, we collect a series of MedISeg tricks for different model implementation phases. We experimentally explore the effectiveness of these tricks on consistent baselines. We also open-sourced a strong MedISeg repository, where each component has the advantage of plug-and-play.
  arXiv  Detail & Related papers  (2022-09-21T12:30:05Z)
• M2Net: Multi-modal Multi-channel Network for Overall Survival Time Prediction of Brain Tumor Patients [151.4352001822956]
  Early and accurate prediction of overall survival (OS) time can help to obtain better treatment planning for brain tumor patients. Existing prediction methods rely on radiomic features at the local lesion area of a magnetic resonance (MR) volume. We propose an end-to-end OS time prediction model; namely, Multi-modal Multi-channel Network (M2Net)
  arXiv  Detail & Related papers  (2020-06-01T05:21:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.