Exploring scalable medical image encoders beyond text supervision

• URL: http://arxiv.org/abs/2401.10815v3
• Date: Fri, 07 Feb 2025 12:03:23 GMT
• Title: Exploring scalable medical image encoders beyond text supervision
• Authors: Fernando Pérez-García, Harshita Sharma, Sam Bond-Taylor, Kenza Bouzid, Valentina Salvatelli, Maximilian Ilse, Shruthi Bannur, Daniel C. Castro, Anton Schwaighofer, Matthew P. Lungren, Maria Teodora Wetscherek, Noel Codella, Stephanie L. Hyland, Javier Alvarez-Valle, Ozan Oktay,
• Abstract summary: Language-supervised pre-training has proven to be a valuable method for extracting semantically meaningful features from images. We introduce RAD-DINO, a biomedical image encoder pre-trained solely on unimodal biomedical imaging data.
• Score: 42.86944965225041
• License:
• Abstract: Language-supervised pre-training has proven to be a valuable method for extracting semantically meaningful features from images, serving as a foundational element in multimodal systems within the computer vision and medical imaging domains. However, the computed features are limited by the information contained in the text, which is particularly problematic in medical imaging, where the findings described by radiologists focus on specific observations. This challenge is compounded by the scarcity of paired imaging-text data due to concerns over leakage of personal health information. In this work, we fundamentally challenge the prevailing reliance on language supervision for learning general-purpose biomedical imaging encoders. We introduce RAD-DINO, a biomedical image encoder pre-trained solely on unimodal biomedical imaging data that obtains similar or greater performance than state-of-the-art biomedical language-supervised models on a diverse range of benchmarks. Specifically, the quality of learned representations is evaluated on standard imaging tasks (classification and semantic segmentation), and a vision-language alignment task (text report generation from images). To further demonstrate the drawback of language supervision, we show that features from RAD-DINO correlate with other medical records (e.g., sex or age) better than language-supervised models, which are generally not mentioned in radiology reports. Finally, we conduct a series of ablations determining the factors in RAD-DINO's performance; notably, we observe that RAD-DINO's downstream performance scales well with the quantity and diversity of training data, demonstrating that image-only supervision is a scalable approach for training a foundational biomedical image encoder. Model weights of RAD-DINO trained on publicly available datasets are available at https://huggingface.co/microsoft/rad-dino.

Related papers

• Expert-level vision-language foundation model for real-world radiology and comprehensive evaluation [27.05259342502574]
  We present RadFound, a vision-language foundation model tailored for radiology. It is trained on the most extensive dataset of over 8.1 million images and 250,000 image-text pairs. To establish expert-level multimodal perception and generation capabilities, RadFound introduces an enhanced vision encoder.
  arXiv  Detail & Related papers  (2024-09-24T15:31:49Z)
• UnSegGNet: Unsupervised Image Segmentation using Graph Neural Networks [9.268228808049951]
  This research contributes to the broader field of unsupervised medical imaging and computer vision. It presents an innovative methodology for image segmentation that aligns with real-world challenges. The proposed method holds promise for diverse applications, including medical imaging, remote sensing, and object recognition.
  arXiv  Detail & Related papers  (2024-05-09T19:02:00Z)
• VALD-MD: Visual Attribution via Latent Diffusion for Medical Diagnostics [0.0]
  Visual attribution in medical imaging seeks to make evident the diagnostically-relevant components of a medical image. We here present a novel generative visual attribution technique, one that leverages latent diffusion models in combination with domain-specific large language models. The resulting system also exhibits a range of latent capabilities including zero-shot localized disease induction.
  arXiv  Detail & Related papers  (2024-01-02T19:51:49Z)
• Radiology Report Generation Using Transformers Conditioned with Non-imaging Data [55.17268696112258]
  This paper proposes a novel multi-modal transformer network that integrates chest x-ray (CXR) images and associated patient demographic information. The proposed network uses a convolutional neural network to extract visual features from CXRs and a transformer-based encoder-decoder network that combines the visual features with semantic text embeddings of patient demographic information.
  arXiv  Detail & Related papers  (2023-11-18T14:52:26Z)
• LVM-Med: Learning Large-Scale Self-Supervised Vision Models for Medical Imaging via Second-order Graph Matching [59.01894976615714]
  We introduce LVM-Med, the first family of deep networks trained on large-scale medical datasets. We have collected approximately 1.3 million medical images from 55 publicly available datasets. LVM-Med empirically outperforms a number of state-of-the-art supervised, self-supervised, and foundation models.
  arXiv  Detail & Related papers  (2023-06-20T22:21:34Z)
• 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 to Exploit Temporal Structure for Biomedical Vision-Language Processing [53.89917396428747]
  Self-supervised learning in vision-language processing exploits semantic alignment between imaging and text modalities. We explicitly account for prior images and reports when available during both training and fine-tuning. Our approach, named BioViL-T, uses a CNN-Transformer hybrid multi-image encoder trained jointly with a text model.
  arXiv  Detail & Related papers  (2023-01-11T16:35:33Z)
• Representative Image Feature Extraction via Contrastive Learning Pretraining for Chest X-ray Report Generation [19.69560434388278]
  The goal of medical report generation is to accurately capture and describe the image findings. Previous works pretrain their visual encoding neural networks with large datasets in different domains. We propose a framework that uses a contrastive learning approach to pretrain the visual encoder and requires no additional meta information.
  arXiv  Detail & Related papers  (2022-09-04T12:07:19Z)
• Self-supervised Multi-modal Training from Uncurated Image and Reports Enables Zero-shot Oversight Artificial Intelligence in Radiology [31.045221580446963]
  We present a model dubbed Medical Cross-attention Vision-Language model (Medical X-VL) Our model enables various zero-shot tasks for oversight AI, ranging from the zero-shot classification to zero-shot error correction. Our method was especially successful in the data-limited setting, suggesting the potential widespread applicability in medical domain.
  arXiv  Detail & Related papers  (2022-08-10T04:35:58Z)
• Auxiliary Signal-Guided Knowledge Encoder-Decoder for Medical Report Generation [107.3538598876467]
  We propose an Auxiliary Signal-Guided Knowledge-Decoder (ASGK) to mimic radiologists' working patterns. ASGK integrates internal visual feature fusion and external medical linguistic information to guide medical knowledge transfer and learning.
  arXiv  Detail & Related papers  (2020-06-06T01:00:15Z)

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.