Source Matters: Source Dataset Impact on Model Robustness in Medical Imaging

• URL: http://arxiv.org/abs/2403.04484v2
• Date: Mon, 19 Aug 2024 13:06:36 GMT
• Title: Source Matters: Source Dataset Impact on Model Robustness in Medical Imaging
• Authors: Dovile Juodelyte, Yucheng Lu, Amelia Jiménez-Sánchez, Sabrina Bottazzi, Enzo Ferrante, Veronika Cheplygina,
• Abstract summary: We show that ImageNet and RadImageNet achieve comparable classification performance. ImageNet is much more prone to overfitting to confounders. We recommend that researchers using ImageNet-pretrained models reexamine their model.
• Score: 14.250975981451914
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Transfer learning has become an essential part of medical imaging classification algorithms, often leveraging ImageNet weights. The domain shift from natural to medical images has prompted alternatives such as RadImageNet, often showing comparable classification performance. However, it remains unclear whether the performance gains from transfer learning stem from improved generalization or shortcut learning. To address this, we conceptualize confounders by introducing the Medical Imaging Contextualized Confounder Taxonomy (MICCAT) and investigate a range of confounders across it -- whether synthetic or sampled from the data -- using two public chest X-ray and CT datasets. We show that ImageNet and RadImageNet achieve comparable classification performance, yet ImageNet is much more prone to overfitting to confounders. We recommend that researchers using ImageNet-pretrained models reexamine their model robustness by conducting similar experiments. Our code and experiments are available at https://github.com/DovileDo/source-matters.

Related papers

• Comparative Analysis of ImageNet Pre-Trained Deep Learning Models and DINOv2 in Medical Imaging Classification [7.205610366609243]
  In this paper, we performed a glioma grading task using three clinical modalities of brain MRI data. We compared the performance of various pre-trained deep learning models, including those based on ImageNet and DINOv2. Our findings indicate that in our clinical dataset, DINOv2's performance was not as strong as ImageNet-based pre-trained models.
  arXiv  Detail & Related papers  (2024-02-12T11:49:08Z)
• Additional Look into GAN-based Augmentation for Deep Learning COVID-19 Image Classification [57.1795052451257]
  We study the dependence of the GAN-based augmentation performance on dataset size with a focus on small samples. We train StyleGAN2-ADA with both sets and then, after validating the quality of generated images, we use trained GANs as one of the augmentations approaches in multi-class classification problems. The GAN-based augmentation approach is found to be comparable with classical augmentation in the case of medium and large datasets but underperforms in the case of smaller datasets.
  arXiv  Detail & Related papers  (2024-01-26T08:28:13Z)
• Performance of GAN-based augmentation for deep learning COVID-19 image classification [57.1795052451257]
  The biggest challenge in the application of deep learning to the medical domain is the availability of training data. Data augmentation is a typical methodology used in machine learning when confronted with a limited data set. In this work, a StyleGAN2-ADA model of Generative Adversarial Networks is trained on the limited COVID-19 chest X-ray image set.
  arXiv  Detail & Related papers  (2023-04-18T15:39:58Z)
• Revisiting Hidden Representations in Transfer Learning for Medical Imaging [2.4545492329339815]
  We compare ImageNet and RadImageNet on seven medical classification tasks. Our results indicate that, contrary to intuition, ImageNet and RadImageNet may converge to distinct intermediate representations. Our findings show that the similarity between networks before and after fine-tuning does not correlate with performance gains.
  arXiv  Detail & Related papers  (2023-02-16T13:04:59Z)
• A Systematic Benchmarking Analysis of Transfer Learning for Medical Image Analysis [7.339428207644444]
  We conduct a systematic study on the transferability of models pre-trained on iNat2021, the most recent large-scale fine-grained dataset. We present a practical approach to bridge the domain gap between natural and medical images by continually (pre-training) supervised ImageNet models on medical images.
  arXiv  Detail & Related papers  (2021-08-12T19:08:34Z)
• Generative Adversarial U-Net for Domain-free Medical Image Augmentation [49.72048151146307]
  The shortage of annotated medical images is one of the biggest challenges in the field of medical image computing. In this paper, we develop a novel generative method named generative adversarial U-Net. Our newly designed model is domain-free and generalizable to various medical images.
  arXiv  Detail & Related papers  (2021-01-12T23:02:26Z)
• Fed-Sim: Federated Simulation for Medical Imaging [131.56325440976207]
  We introduce a physics-driven generative approach that consists of two learnable neural modules. We show that our data synthesis framework improves the downstream segmentation performance on several datasets.
  arXiv  Detail & Related papers  (2020-09-01T19:17:46Z)
• Comparing to Learn: Surpassing ImageNet Pretraining on Radiographs By Comparing Image Representations [39.08296644280442]
  We propose a new pretraining method which learns from 700k radiographs given no manual annotations. We call our method as Comparing to Learn (C2L) because it learns robust features by comparing different image representations. The experimental results on radiographs show that C2L can outperform ImageNet pretraining and previous state-of-the-art approaches significantly.
  arXiv  Detail & Related papers  (2020-07-15T01:14:34Z)
• From ImageNet to Image Classification: Contextualizing Progress on Benchmarks [99.19183528305598]
  We study how specific design choices in the ImageNet creation process impact the fidelity of the resulting dataset. Our analysis pinpoints how a noisy data collection pipeline can lead to a systematic misalignment between the resulting benchmark and the real-world task it serves as a proxy for.
  arXiv  Detail & Related papers  (2020-05-22T17:39:16Z)
• I Am Going MAD: Maximum Discrepancy Competition for Comparing Classifiers Adaptively [135.7695909882746]
  We name the MAximum Discrepancy (MAD) competition. We adaptively sample a small test set from an arbitrarily large corpus of unlabeled images. Human labeling on the resulting model-dependent image sets reveals the relative performance of the competing classifiers.
  arXiv  Detail & Related papers  (2020-02-25T03:32:29Z)

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.