Can Medical Vision-Language Pre-training Succeed with Purely Synthetic Data?

• URL: http://arxiv.org/abs/2410.13523v1
• Date: Thu, 17 Oct 2024 13:11:07 GMT
• Title: Can Medical Vision-Language Pre-training Succeed with Purely Synthetic Data?
• Authors: Che Liu, Zhongwei Wan, Haozhe Wang, Yinda Chen, Talha Qaiser, Chen Jin, Fariba Yousefi, Nikolay Burlutskiy, Rossella Arcucci,
• Abstract summary: Training medical vision-language pre-training models requires datasets with paired, high-quality image-text data. Recent advancements in Large Language Models have made it possible to generate large-scale synthetic image-text pairs. We propose an automated pipeline to build a diverse, high-quality synthetic dataset.
• Score: 8.775988650381397
• License:
• Abstract: Medical Vision-Language Pre-training (MedVLP) has made significant progress in enabling zero-shot tasks for medical image understanding. However, training MedVLP models typically requires large-scale datasets with paired, high-quality image-text data, which are scarce in the medical domain. Recent advancements in Large Language Models (LLMs) and diffusion models have made it possible to generate large-scale synthetic image-text pairs. This raises the question: *Can MedVLP succeed using purely synthetic data?* To address this, we use off-the-shelf generative models to create synthetic radiology reports and paired Chest X-ray (CXR) images, and propose an automated pipeline to build a diverse, high-quality synthetic dataset, enabling a rigorous study that isolates model and training settings, focusing entirely from the data perspective. Our results show that MedVLP models trained *exclusively on synthetic data* outperform those trained on real data by **3.8%** in averaged AUC on zero-shot classification. Moreover, using a combination of synthetic and real data leads to a further improvement of **9.07%**. Additionally, MedVLP models trained on synthetic or mixed data consistently outperform those trained on real data in zero-shot grounding, as well as in fine-tuned classification and segmentation tasks. Our analysis suggests MedVLP trained on well-designed synthetic data can outperform models trained on real datasets, which may be limited by low-quality samples and long-tailed distributions.

Related papers

• Data-Constrained Synthesis of Training Data for De-Identification [0.0]
  We domain-adapt large language models (LLMs) to the clinical domain. We generate synthetic clinical texts that are machine-annotated with tags for personally identifiable information. The synthetic corpora are then used to train synthetic NER models.
  arXiv  Detail & Related papers  (2025-02-20T16:09:27Z)
• License Plate Images Generation with Diffusion Models [4.151073288078749]
  We propose to synthesize realistic license plates (LPs) using diffusion models inspired by video generation. We have prepared a synthetic dataset consisting of 10,000 LP images, publicly available at https://zenodo.org/doi/10.528/zenodo.13342102.
  arXiv  Detail & Related papers  (2025-01-06T20:22:18Z)
• Embryo 2.0: Merging Synthetic and Real Data for Advanced AI Predictions [69.07284335967019]
  We train two generative models using two datasets, one created and made publicly available, and one existing public dataset. We generate synthetic embryo images at various cell stages, including 2-cell, 4-cell, 8-cell, morula, and blastocyst. These were combined with real images to train classification models for embryo cell stage prediction.
  arXiv  Detail & Related papers  (2024-12-02T08:24:49Z)
• Chatting Up Attachment: Using LLMs to Predict Adult Bonds [0.0]
  We use GPT-4 and Claude 3 Opus to create agents that simulate adults with varying profiles, childhood memories, and attachment styles. We evaluate our models using a transcript dataset from 9 humans who underwent the same interview protocol, analyzed and labeled by mental health professionals. Our findings indicate that training the models using only synthetic data achieves performance comparable to training the models on human data.
  arXiv  Detail & Related papers  (2024-08-31T04:29:19Z)
• Learning from Synthetic Data for Visual Grounding [55.21937116752679]
  We show that SynGround can improve the localization capabilities of off-the-shelf vision-and-language models. Data generated with SynGround improves the pointing game accuracy of a pretrained ALBEF and BLIP models by 4.81% and 17.11% absolute percentage points, respectively.
  arXiv  Detail & Related papers  (2024-03-20T17:59:43Z)
• Synthetically Enhanced: Unveiling Synthetic Data's Potential in Medical Imaging Research [4.475998415951477]
  Generative AI offers a promising approach to generating synthetic images, enhancing dataset diversity. This study investigates the impact of synthetic data supplementation on the performance and generalizability of medical imaging research.
  arXiv  Detail & Related papers  (2023-11-15T21:58:01Z)
• Let's Synthesize Step by Step: Iterative Dataset Synthesis with Large Language Models by Extrapolating Errors from Small Models [69.76066070227452]
  *Data Synthesis* is a promising way to train a small model with very little labeled data. We propose *Synthesis Step by Step* (**S3**), a data synthesis framework that shrinks this distribution gap. Our approach improves the performance of a small model by reducing the gap between the synthetic dataset and the real data.
  arXiv  Detail & Related papers  (2023-10-20T17:14:25Z)
• How Good Are Synthetic Medical Images? An Empirical Study with Lung Ultrasound [0.3312417881789094]
  Adding synthetic training data using generative models offers a low-cost method to deal with the data scarcity challenge. We show that training with both synthetic and real data outperforms training with real data alone.
  arXiv  Detail & Related papers  (2023-10-05T15:42:53Z)
• Synthetic data, real errors: how (not) to publish and use synthetic data [86.65594304109567]
  We show how the generative process affects the downstream ML task. We introduce Deep Generative Ensemble (DGE) to approximate the posterior distribution over the generative process model parameters.
  arXiv  Detail & Related papers  (2023-05-16T07:30:29Z)
• 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)
• PromDA: Prompt-based Data Augmentation for Low-Resource NLU Tasks [61.51515750218049]
  This paper focuses on the Data Augmentation for low-resource Natural Language Understanding (NLU) tasks. We propose Prompt-based Data Augmentation model (PromDA) which only trains small-scale Soft Prompt. PromDA generates synthetic data via two different views and filters out the low-quality data using NLU models.
  arXiv  Detail & Related papers  (2022-02-25T05:09:27Z)

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.