General Methods Make Great Domain-specific Foundation Models: A Case-study on Fetal Ultrasound

• URL: http://arxiv.org/abs/2506.19552v1
• Date: Tue, 24 Jun 2025 12:00:13 GMT
• Title: General Methods Make Great Domain-specific Foundation Models: A Case-study on Fetal Ultrasound
• Authors: Jakob Ambsdorf, Asbjørn Munk, Sebastian Llambias, Anders Nymark Christensen, Kamil Mikolaj, Randall Balestriero, Martin Tolsgaard, Aasa Feragen, Mads Nielsen,
• Abstract summary: We train a foundation model on a large regional fetal ultrasound dataset of 2M images.<n>We compare against a series of models pretrained on natural images, ultrasound images, and supervised baselines.
• Score: 15.08091031875334
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: With access to large-scale, unlabeled medical datasets, researchers are confronted with two questions: Should they attempt to pretrain a custom foundation model on this medical data, or use transfer-learning from an existing generalist model? And, if a custom model is pretrained, are novel methods required? In this paper we explore these questions by conducting a case-study, in which we train a foundation model on a large regional fetal ultrasound dataset of 2M images. By selecting the well-established DINOv2 method for pretraining, we achieve state-of-the-art results on three fetal ultrasound datasets, covering data from different countries, classification, segmentation, and few-shot tasks. We compare against a series of models pretrained on natural images, ultrasound images, and supervised baselines. Our results demonstrate two key insights: (i) Pretraining on custom data is worth it, even if smaller models are trained on less data, as scaling in natural image pretraining does not translate to ultrasound performance. (ii) Well-tuned methods from computer vision are making it feasible to train custom foundation models for a given medical domain, requiring no hyperparameter tuning and little methodological adaptation. Given these findings, we argue that a bias towards methodological innovation should be avoided when developing domain specific foundation models under common computational resource constraints.

Related papers

• Weakly supervised deep learning model with size constraint for prostate cancer detection in multiparametric MRI and generalization to unseen domains [0.90668179713299]
  We show that the model achieves on-par performance with strong fully supervised baseline models. We also observe a performance decrease for both fully supervised and weakly supervised models when tested on unseen data domains.
  arXiv  Detail & Related papers  (2024-11-04T12:24:33Z)
• How Good Are We? Evaluating Cell AI Foundation Models in Kidney Pathology with Human-in-the-Loop Enrichment [11.60167559546617]
  Training AI foundation models have emerged as a promising large-scale learning approach for addressing real-world healthcare challenges. While many of these models have been developed for tasks like disease diagnosis and tissue quantification, their readiness for deployment on some arguably simplest tasks, such as nuclei segmentation within a single organ, remains uncertain. This paper seeks to answer this key question, "How good are we?" by thoroughly evaluating the performance of recent cell foundation models on a curated dataset.
  arXiv  Detail & Related papers  (2024-10-31T17:00:33Z)
• Text-guided Foundation Model Adaptation for Long-Tailed Medical Image Classification [4.6651139122498]
  In medical contexts, the imbalanced data distribution in long-tailed datasets, due to scarce labels for rare diseases, greatly impairs the diagnostic accuracy of deep learning models. Recent multimodal text-image supervised foundation models offer new solutions to data scarcity through effective representation learning. We propose a novel Text-guided Foundation model Adaptation for Long-Tailed medical image classification (TFA-LT) Our method achieves an accuracy improvement of up to 27.1%, highlighting the substantial potential of foundation model adaptation in this area.
  arXiv  Detail & Related papers  (2024-08-27T04:18:18Z)
• Overcoming Data Scarcity in Biomedical Imaging with a Foundational Multi-Task Model [2.5994154212235685]
  Foundational models, pretrained on a large scale, have demonstrated substantial success across non-medical domains. Here, we propose a multi-task learning strategy that decouples the number of training tasks from memory requirements.
  arXiv  Detail & Related papers  (2023-11-16T12:20:25Z)
• Interpretable 2D Vision Models for 3D Medical Images [47.75089895500738]
  This study proposes a simple approach of adapting 2D networks with an intermediate feature representation for processing 3D images. We show on all 3D MedMNIST datasets as benchmark and two real-world datasets consisting of several hundred high-resolution CT or MRI scans that our approach performs on par with existing methods.
  arXiv  Detail & Related papers  (2023-07-13T08:27:09Z)
• 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)
• Prototype Learning for Explainable Brain Age Prediction [1.104960878651584]
  We present ExPeRT, an explainable prototype-based model specifically designed for regression tasks. Our proposed model makes a sample prediction from the distances to a set of learned prototypes in latent space, using a weighted mean of prototype labels. Our approach achieved state-of-the-art prediction performance while providing insight into the model's reasoning process.
  arXiv  Detail & Related papers  (2023-06-16T14:13:21Z)
• Universal Domain Adaptation from Foundation Models: A Baseline Study [58.51162198585434]
  We make empirical studies of state-of-the-art UniDA methods using foundation models. We introduce textitCLIP distillation, a parameter-free method specifically designed to distill target knowledge from CLIP models. Although simple, our method outperforms previous approaches in most benchmark tasks.
  arXiv  Detail & Related papers  (2023-05-18T16:28:29Z)
• 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)
• Synthetic Model Combination: An Instance-wise Approach to Unsupervised Ensemble Learning [92.89846887298852]
  Consider making a prediction over new test data without any opportunity to learn from a training set of labelled data. Give access to a set of expert models and their predictions alongside some limited information about the dataset used to train them.
  arXiv  Detail & Related papers  (2022-10-11T10:20:31Z)
• Ultrasound Signal Processing: From Models to Deep Learning [64.56774869055826]
  Medical ultrasound imaging relies heavily on high-quality signal processing to provide reliable and interpretable image reconstructions. Deep learning based methods, which are optimized in a data-driven fashion, have gained popularity. A relatively new paradigm combines the power of the two: leveraging data-driven deep learning, as well as exploiting domain knowledge.
  arXiv  Detail & Related papers  (2022-04-09T13:04:36Z)
• On-the-Fly Test-time Adaptation for Medical Image Segmentation [63.476899335138164]
  Adapting the source model to target data distribution at test-time is an efficient solution for the data-shift problem. We propose a new framework called Adaptive UNet where each convolutional block is equipped with an adaptive batch normalization layer. During test-time, the model takes in just the new test image and generates a domain code to adapt the features of source model according to the test data.
  arXiv  Detail & Related papers  (2022-03-10T18:51:29Z)

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.