Learning Robust Visual Features in Computed Tomography Enables Efficient Transfer Learning for Clinical Tasks

• URL: http://arxiv.org/abs/2604.04133v1
• Date: Sun, 05 Apr 2026 14:29:35 GMT
• Title: Learning Robust Visual Features in Computed Tomography Enables Efficient Transfer Learning for Clinical Tasks
• Abstract summary: We present VoxelFM, a 3D CT foundation model trained with self-distillation using the DINO framework.<n>We evaluate VoxelFM across seven categories of clinically relevant downstream tasks using frozen backbone representations with lightweight probes.
• Score: 6.299805867949439
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: There is substantial interest in developing artificial intelligence systems to support radiologists across tasks ranging from segmentation to report generation. Existing computed tomography (CT) foundation models have largely focused on building generalist vision-language systems capable of tasks such as question answering and report generation. However, training reliable vision-language systems requires paired image-text data at a scale that remains unavailable in CT. Moreover, adapting the underlying visual representations to downstream tasks typically requires partial or full backbone fine-tuning, a computationally demanding process inaccessible to many research groups. Instead, foundation models should prioritise learning robust visual representations that enable efficient transfer to new tasks with minimal labelled data and without backbone fine-tuning. We present VoxelFM, a 3D CT foundation model trained with self-distillation using the DINO framework, which learns semantically rich features without language supervision. We evaluated VoxelFM across seven categories of clinically relevant downstream tasks using frozen backbone representations with lightweight probes: classification, regression, survival analysis, instance retrieval, localisation, segmentation, and report generation. VoxelFM matched or outperformed four existing CT foundation models across all task categories. Despite receiving no language supervision during pre-training, VoxelFM surpassed models explicitly trained with language-alignment objectives, including on report generation. Our results indicate that current CT foundation models perform significantly better as feature extractors for lightweight probes rather than as vision encoders for vision-language models. Model weights and training code are publicly available.

Related papers

• Foundation Model for Skeleton-Based Human Action Understanding [56.89025287217221]
  This paper presents a Unified Skeleton-based Dense Representation Learning framework.<n>USDRL consists of a Transformer-based Dense Spatio-Temporal (DSTE), Multi-Grained Feature Decorrelation (MG-FD), and Multi-Perspective Consistency Training (MPCT)
  arXiv  Detail & Related papers  (2025-08-18T02:42:16Z)
• Evaluating Vision Language Models (VLMs) for Radiology: A Comprehensive Analysis [4.803310914375717]
  This study evaluates three vision-language foundation models (RAD-DINO, CheXagent, and BiomedCLIP) on their ability to capture fine-grained imaging features for radiology tasks.<n>The models were assessed across classification, segmentation, and regression tasks for pneumothorax and cardiomegaly on chest radiographs.
  arXiv  Detail & Related papers  (2025-04-22T17:20:34Z)
• Towards Universal Text-driven CT Image Segmentation [4.76971404389011]
  We propose OpenVocabCT, a vision-language model pretrained on large-scale 3D CT images for universal text-driven segmentation.<n>We decompose the diagnostic reports into fine-grained, organ-level descriptions using large language models for multi-granular contrastive learning.
  arXiv  Detail & Related papers  (2025-03-08T03:02:57Z)
• Fake It Till You Make It: Using Synthetic Data and Domain Knowledge for Improved Text-Based Learning for LGE Detection [11.532639713283226]
  We use strategies rooted in domain knowledge to train a model for LGE detection using text from clinical reports.<n>We standardize the orientation of the images in an anatomy-informed way to enable better alignment of spatial and text features.<n> ablation studies are carried out to elucidate the contributions of each design component to the overall performance of the model.
  arXiv  Detail & Related papers  (2025-02-18T15:30:48Z)
• Medical Vision-Language Pre-Training for Brain Abnormalities [96.1408455065347]
  We show how to automatically collect medical image-text aligned data for pretraining from public resources such as PubMed. In particular, we present a pipeline that streamlines the pre-training process by initially collecting a large brain image-text dataset. We also investigate the unique challenge of mapping subfigures to subcaptions in the medical domain.
  arXiv  Detail & Related papers  (2024-04-27T05:03:42Z)
• See Through Their Minds: Learning Transferable Neural Representation from Cross-Subject fMRI [32.40827290083577]
  Deciphering visual content from functional Magnetic Resonance Imaging (fMRI) helps illuminate the human vision system. Previous approaches primarily employ subject-specific models, sensitive to training sample size. We propose shallow subject-specific adapters to map cross-subject fMRI data into unified representations. During training, we leverage both visual and textual supervision for multi-modal brain decoding.
  arXiv  Detail & Related papers  (2024-03-11T01:18:49Z)
• Data-efficient Large Vision Models through Sequential Autoregression [58.26179273091461]
  We develop an efficient, autoregression-based vision model on a limited dataset. We demonstrate how this model achieves proficiency in a spectrum of visual tasks spanning both high-level and low-level semantic understanding. Our empirical evaluations underscore the model's agility in adapting to various tasks, heralding a significant reduction in the parameter footprint.
  arXiv  Detail & Related papers  (2024-02-07T13:41:53Z)
• Visual Grounding Helps Learn Word Meanings in Low-Data Regimes [47.7950860342515]
  Modern neural language models (LMs) are powerful tools for modeling human sentence production and comprehension. But to achieve these results, LMs must be trained in distinctly un-human-like ways. Do models trained more naturalistically -- with grounded supervision -- exhibit more humanlike language learning? We investigate this question in the context of word learning, a key sub-task in language acquisition.
  arXiv  Detail & Related papers  (2023-10-20T03:33:36Z)
• Vision-Language Modelling For Radiological Imaging and Reports In The Low Data Regime [70.04389979779195]
  This paper explores training medical vision-language models (VLMs) where the visual and language inputs are embedded into a common space. We explore several candidate methods to improve low-data performance, including adapting generic pre-trained models to novel image and text domains. Using text-to-image retrieval as a benchmark, we evaluate the performance of these methods with variable sized training datasets of paired chest X-rays and radiological reports.
  arXiv  Detail & Related papers  (2023-03-30T18:20:00Z)
• Learning from Temporal Spatial Cubism for Cross-Dataset Skeleton-based Action Recognition [88.34182299496074]
  Action labels are only available on a source dataset, but unavailable on a target dataset in the training stage. We utilize a self-supervision scheme to reduce the domain shift between two skeleton-based action datasets. By segmenting and permuting temporal segments or human body parts, we design two self-supervised learning classification tasks.
  arXiv  Detail & Related papers  (2022-07-17T07:05:39Z)
• Revisiting Classifier: Transferring Vision-Language Models for Video Recognition [102.93524173258487]
  Transferring knowledge from task-agnostic pre-trained deep models for downstream tasks is an important topic in computer vision research. In this study, we focus on transferring knowledge for video classification tasks. We utilize the well-pretrained language model to generate good semantic target for efficient transferring learning.
  arXiv  Detail & Related papers  (2022-07-04T10:00:47Z)
• X-Learner: Learning Cross Sources and Tasks for Universal Visual Representation [71.51719469058666]
  We propose a representation learning framework called X-Learner. X-Learner learns the universal feature of multiple vision tasks supervised by various sources. X-Learner achieves strong performance on different tasks without extra annotations, modalities and computational costs.
  arXiv  Detail & Related papers  (2022-03-16T17:23:26Z)

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.