Towards Scalable Language-Image Pre-training for 3D Medical Imaging

• URL: http://arxiv.org/abs/2505.21862v1
• Date: Wed, 28 May 2025 01:16:34 GMT
• Title: Towards Scalable Language-Image Pre-training for 3D Medical Imaging
• Authors: Chenhui Zhao, Yiwei Lyu, Asadur Chowdury, Edward Harake, Akhil Kondepudi, Akshay Rao, Xinhai Hou, Honglak Lee, Todd Hollon,
• Abstract summary: We introduce Hierarchical attention for Language-Image Pre-training (HLIP), a scalable pre-training framework for 3D medical imaging.<n>HLIP adopts a lightweight hierarchical attention mechanism inspired by the natural hierarchy of radiology data: slice, scan, and study.<n>Trained on 220K patients with 3.13 million scans for brain MRI and 240K patients with 1.44 million scans for head CT, HLIP achieves state-of-the-art performance.
• Score: 49.18894445671976
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Language-image pre-training has demonstrated strong performance in 2D medical imaging, but its success in 3D modalities such as CT and MRI remains limited due to the high computational demands of volumetric data, which pose a significant barrier to training on large-scale, uncurated clinical studies. In this study, we introduce Hierarchical attention for Language-Image Pre-training (HLIP), a scalable pre-training framework for 3D medical imaging. HLIP adopts a lightweight hierarchical attention mechanism inspired by the natural hierarchy of radiology data: slice, scan, and study. This mechanism exhibits strong generalizability, e.g., +4.3% macro AUC on the Rad-ChestCT benchmark when pre-trained on CT-RATE. Moreover, the computational efficiency of HLIP enables direct training on uncurated datasets. Trained on 220K patients with 3.13 million scans for brain MRI and 240K patients with 1.44 million scans for head CT, HLIP achieves state-of-the-art performance, e.g., +32.4% balanced ACC on the proposed publicly available brain MRI benchmark Pub-Brain-5; +1.4% and +6.9% macro AUC on head CT benchmarks RSNA and CQ500, respectively. These results demonstrate that, with HLIP, directly pre-training on uncurated clinical datasets is a scalable and effective direction for language-image pre-training in 3D medical imaging. The code is available at https://github.com/Zch0414/hlip

Related papers

• Triad: Vision Foundation Model for 3D Magnetic Resonance Imaging [3.7942449131350413]
  We propose Triad, a vision foundation model for 3D MRI.<n> Triad adopts a widely used autoencoder architecture to learn robust representations from 131,170 3D MRI volumes.<n>We evaluate Triad across three tasks, namely, organ/tumor segmentation, organ/cancer classification, and medical image registration.
  arXiv  Detail & Related papers  (2025-02-19T19:31:52Z)
• CT-GLIP: 3D Grounded Language-Image Pretraining with CT Scans and Radiology Reports for Full-Body Scenarios [53.94122089629544]
  We introduce CT-GLIP (Grounded Language-Image Pretraining with CT scans), a novel method that constructs organ-level image-text pairs to enhance multimodal contrastive learning. Our method, trained on a multimodal CT dataset comprising 44,011 organ-level vision-text pairs from 17,702 patients across 104 organs, demonstrates it can identify organs and abnormalities in a zero-shot manner using natural languages.
  arXiv  Detail & Related papers  (2024-04-23T17:59:01Z)
• Developing Generalist Foundation Models from a Multimodal Dataset for 3D Computed Tomography [10.110878689623961]
  We introduce CT-RATE, the first dataset that pairs 3D medical images with corresponding textual reports.<n>We develop CT-CLIP, a CT-focused contrastive language-image pretraining framework.<n>We create CT-CHAT, a vision-language foundational chat model for 3D chest CT volumes.
  arXiv  Detail & Related papers  (2024-03-26T16:19:56Z)
• Disruptive Autoencoders: Leveraging Low-level features for 3D Medical Image Pre-training [51.16994853817024]
  This work focuses on designing an effective pre-training framework for 3D radiology images. We introduce Disruptive Autoencoders, a pre-training framework that attempts to reconstruct the original image from disruptions created by a combination of local masking and low-level perturbations. The proposed pre-training framework is tested across multiple downstream tasks and achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2023-07-31T17:59:42Z)
• 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)
• Video4MRI: An Empirical Study on Brain Magnetic Resonance Image Analytics with CNN-based Video Classification Frameworks [60.42012344842292]
  3D CNN-based models dominate the field of magnetic resonance image (MRI) analytics. In this paper, four datasets of Alzheimer's and Parkinson's disease recognition are utilized in experiments. In terms of efficiency, the video framework performs better than 3D-CNN models by 5% - 11% with 50% - 66% less trainable parameters.
  arXiv  Detail & Related papers  (2023-02-24T15:26:31Z)
• Slice-level Detection of Intracranial Hemorrhage on CT Using Deep Descriptors of Adjacent Slices [0.31317409221921133]
  We propose a new strategy to train emphslice-level classifiers on CT scans based on the descriptors of the adjacent slices along the axis. We obtain a single model in the top 4% best-performing solutions of the RSNA Intracranial Hemorrhage dataset challenge. The proposed method is general and can be applied to other 3D medical diagnosis tasks such as MRI imaging.
  arXiv  Detail & Related papers  (2022-08-05T23:20:37Z)
• Self-supervised 3D anatomy segmentation using self-distilled masked image transformer (SMIT) [2.7298989068857487]
  Self-supervised learning has demonstrated success in medical image segmentation using convolutional networks. We show our approach is more accurate and requires fewer fine tuning datasets than other pretext tasks.
  arXiv  Detail & Related papers  (2022-05-20T17:55:14Z)
• Medical Transformer: Universal Brain Encoder for 3D MRI Analysis [1.6287500717172143]
  Existing 3D-based methods have transferred the pre-trained models to downstream tasks. They demand a massive amount of parameters to train the model for 3D medical imaging. We propose a novel transfer learning framework, called Medical Transformer, that effectively models 3D volumetric images in the form of a sequence of 2D image slices.
  arXiv  Detail & Related papers  (2021-04-28T08:34:21Z)
• Deep Q-Network-Driven Catheter Segmentation in 3D US by Hybrid Constrained Semi-Supervised Learning and Dual-UNet [74.22397862400177]
  We propose a novel catheter segmentation approach, which requests fewer annotations than the supervised learning method. Our scheme considers a deep Q learning as the pre-localization step, which avoids voxel-level annotation. With the detected catheter, patch-based Dual-UNet is applied to segment the catheter in 3D volumetric data.
  arXiv  Detail & Related papers  (2020-06-25T21:10:04Z)

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.