OmniRad: A Radiological Foundation Model for Multi-Task Medical Image Analysis

• URL: http://arxiv.org/abs/2602.04547v1
• Date: Wed, 04 Feb 2026 13:38:51 GMT
• Title: OmniRad: A Radiological Foundation Model for Multi-Task Medical Image Analysis
• Authors: Luca Zedda, Andrea Loddo, Cecilia Di Ruberto,
• Abstract summary: We introduce OmniRad, a self-supervised foundation model pretrained on 1.2 million medical images.<n>We evaluate it on a broad suite of public benchmarks spanning classification and segmentation across multiple modalities.
• Score: 2.8826431001526616
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Radiological analysis increasingly benefits from pretrained visual representations that can support heterogeneous downstream tasks across imaging modalities. In this work, we introduce OmniRad, a self-supervised radiological foundation model pretrained on 1.2 million medical images, designed with radiology-inspired principles emphasizing representation reuse and cross-task transferability. We evaluate the pretrained encoder under multiple downstream adaptation regimes, including lightweight task-specific adapters with a frozen backbone as well as full end-to-end fine-tuning for classification, allowing us to assess both representation quality and task-specific performance. OmniRad is evaluated on a broad suite of public benchmarks spanning classification and segmentation across multiple modalities. On the MedMNISTv2 collection, OmniRad improves classification F1 by up to 2.05% over competing foundation models. For dense prediction, OmniRad attains mean Dice score improvements across six MedSegBench datasets when using frozen representations. Qualitative analyses and latent-space visualizations suggest improved feature clustering and modality-related separation.

Related papers

• UAM: A Unified Attention-Mamba Backbone of Multimodal Framework for Tumor Cell Classification [1.529342790344802]
  We introduce a Unified Attention-Mamba backbone for cell-level classification using radiomics features.<n>We propose a multimodal UAM framework that jointly performs cell-level classification and image segmentation.
  arXiv  Detail & Related papers  (2025-11-21T16:18:55Z)
• Multivariate Gaussian Representation Learning for Medical Action Evaluation [6.117273466254055]
  We introduce CPRE-6k, a multi-temporalview, multi-label medical action benchmark containing 6,372 expert-valnotated videos with 22 clinical labels.<n>We presentssAct, a framework to advance medical motion analysis through temporaltemporal learning.
  arXiv  Detail & Related papers  (2025-11-13T08:01:58Z)
• A Fully Open and Generalizable Foundation Model for Ultrasound Clinical Applications [77.3888788549565]
  We present EchoCare, a novel ultrasound foundation model for generalist clinical use.<n>We developed EchoCare via self-supervised learning on our curated, publicly available, large-scale dataset EchoCareData.<n>With minimal training, EchoCare outperforms state-of-the-art comparison models across 10 representative ultrasound benchmarks.
  arXiv  Detail & Related papers  (2025-09-15T10:05:31Z)
• SimCroP: Radiograph Representation Learning with Similarity-driven Cross-granularity Pre-training [25.763109982379703]
  We propose a Similarity-Driven Cross-Granularity Pre-training framework on chest CTs.<n>It combines similarity-driven alignment and cross-granularity fusion to improve radiograph interpretation.<n>SimCroP is pre-trained on a large-scale paired CT-reports dataset and validated on image classification and segmentation tasks.
  arXiv  Detail & Related papers  (2025-09-10T06:20:53Z)
• Adapting Foundation Model for Dental Caries Detection with Dual-View Co-Training [53.77904429789069]
  We present Attention-TNet, a novel Dual-View Co-Training network for accurate dental caries detection.<n>OurTNet starts with employing automated tooth detection to establish two complementary views: a global view from panoramic X-ray images and a local view from cropped tooth images.<n>To effectively integrate information from both views, we introduce a Gated Cross-View module.
  arXiv  Detail & Related papers  (2025-08-28T14:13:26Z)
• Large Kernel MedNeXt for Breast Tumor Segmentation and Self-Normalizing Network for pCR Classification in Magnetic Resonance Images [0.0]
  We employ a large- kernel MedNeXt architecture with a two-stage training strategy that expands the receptive field from 3x3x3 to 5x5x5 kernels.<n>For pCR classification, we trained a self-normalizing network (SNN) on radiomic features extracted from the predicted segmentations.<n>Our findings highlight the benefits of combining larger receptive fields and radiomics-driven classification.
  arXiv  Detail & Related papers  (2025-08-03T16:37:14Z)
• AuxDet: Auxiliary Metadata Matters for Omni-Domain Infrared Small Target Detection [49.81255045696323]
  We present the Auxiliary Metadata Driven Infrared Small Target Detector (AuxDet)<n>AuxDet integrates metadata semantics with visual features, guiding adaptive representation learning for each sample.<n>Experiments on the challenging WideIRSTD-Full benchmark demonstrate that AuxDet consistently outperforms state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-05-21T07:02:05Z)
• Generalizing Medical Image Representations via Quaternion Wavelet Networks [9.836302410524842]
  We introduce a novel, generalizable, data- and task-agnostic framework able to extract salient features from medical images.<n>The proposed quaternion wavelet network (QUAVE) can be easily integrated with any pre-existing medical image analysis or synthesis task.
  arXiv  Detail & Related papers  (2023-10-16T09:34:06Z)
• Revisiting the Evaluation of Image Synthesis with GANs [55.72247435112475]
  This study presents an empirical investigation into the evaluation of synthesis performance, with generative adversarial networks (GANs) as a representative of generative models. In particular, we make in-depth analyses of various factors, including how to represent a data point in the representation space, how to calculate a fair distance using selected samples, and how many instances to use from each set.
  arXiv  Detail & Related papers  (2023-04-04T17:54:32Z)
• Improving Classification Model Performance on Chest X-Rays through Lung Segmentation [63.45024974079371]
  We propose a deep learning approach to enhance abnormal chest x-ray (CXR) identification performance through segmentations. Our approach is designed in a cascaded manner and incorporates two modules: a deep neural network with criss-cross attention modules (XLSor) for localizing lung region in CXR images and a CXR classification model with a backbone of a self-supervised momentum contrast (MoCo) model pre-trained on large-scale CXR data sets.
  arXiv  Detail & Related papers  (2022-02-22T15:24:06Z)
• Cross-Modal Contrastive Learning for Abnormality Classification and Localization in Chest X-rays with Radiomics using a Feedback Loop [63.81818077092879]
  We propose an end-to-end semi-supervised cross-modal contrastive learning framework for medical images. We first apply an image encoder to classify the chest X-rays and to generate the image features. The radiomic features are then passed through another dedicated encoder to act as the positive sample for the image features generated from the same chest X-ray.
  arXiv  Detail & Related papers  (2021-04-11T09:16:29Z)

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.