CROCODILE: Causality aids RObustness via COntrastive DIsentangled LEarning
- URL: http://arxiv.org/abs/2408.04949v1
- Date: Fri, 9 Aug 2024 09:08:06 GMT
- Title: CROCODILE: Causality aids RObustness via COntrastive DIsentangled LEarning
- Authors: Gianluca Carloni, Sotirios A Tsaftaris, Sara Colantonio,
- Abstract summary: We introduce our CROCODILE framework, showing how tools from causality can foster a model's robustness to domain shift.
We apply our method to multi-label lung disease classification from CXRs, utilizing over 750000 images.
- Score: 8.975676404678374
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Due to domain shift, deep learning image classifiers perform poorly when applied to a domain different from the training one. For instance, a classifier trained on chest X-ray (CXR) images from one hospital may not generalize to images from another hospital due to variations in scanner settings or patient characteristics. In this paper, we introduce our CROCODILE framework, showing how tools from causality can foster a model's robustness to domain shift via feature disentanglement, contrastive learning losses, and the injection of prior knowledge. This way, the model relies less on spurious correlations, learns the mechanism bringing from images to prediction better, and outperforms baselines on out-of-distribution (OOD) data. We apply our method to multi-label lung disease classification from CXRs, utilizing over 750000 images from four datasets. Our bias-mitigation method improves domain generalization and fairness, broadening the applicability and reliability of deep learning models for a safer medical image analysis. Find our code at: https://github.com/gianlucarloni/crocodile.
Related papers
- An analysis of data variation and bias in image-based dermatological datasets for machine learning classification [2.039829968340841]
In clinical dermatology, classification models can detect malignant lesions on patients' skin using only RGB images as input.
Most learning-based methods employ data acquired from dermoscopic datasets on training, which are large and validated by a gold standard.
This work aims to evaluate the gap between dermoscopic and clinical samples and understand how the dataset variations impact training.
arXiv Detail & Related papers (2025-01-15T17:18:46Z) - Mask of truth: model sensitivity to unexpected regions of medical images [0.9896218845636701]
We challenge the capacity of convolutional neural networks (CNN) to classify chest X-rays and eye fundus images.
We show that all models trained on the Pad dataset, irrespective of the masking strategy, are able to obtain an Area Under the Curve (AUC) above random.
We also reveal a possible spurious correlation in the Chaksu dataset while the performances are more aligned with the expectation of an unbiased model.
arXiv Detail & Related papers (2024-12-05T10:06:58Z) - How Does Pruning Impact Long-Tailed Multi-Label Medical Image
Classifiers? [49.35105290167996]
Pruning has emerged as a powerful technique for compressing deep neural networks, reducing memory usage and inference time without significantly affecting overall performance.
This work represents a first step toward understanding the impact of pruning on model behavior in deep long-tailed, multi-label medical image classification.
arXiv Detail & Related papers (2023-08-17T20:40:30Z) - Chest X-ray Image Classification: A Causal Perspective [49.87607548975686]
We propose a causal approach to address the CXR classification problem, which constructs a structural causal model (SCM) and uses the backdoor adjustment to select effective visual information for CXR classification.
Experimental results demonstrate that our proposed method outperforms the open-source NIH ChestX-ray14 in terms of classification performance.
arXiv Detail & Related papers (2023-05-20T03:17:44Z) - Realistic Data Enrichment for Robust Image Segmentation in
Histopathology [2.248423960136122]
We propose a new approach, based on diffusion models, which can enrich an imbalanced dataset with plausible examples from underrepresented groups.
Our method can simply expand limited clinical datasets making them suitable to train machine learning pipelines.
arXiv Detail & Related papers (2023-04-19T09:52:50Z) - 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) - RadTex: Learning Efficient Radiograph Representations from Text Reports [7.090896766922791]
We build a data-efficient learning framework that utilizes radiology reports to improve medical image classification performance with limited labeled data.
Our model achieves higher classification performance than ImageNet-supervised pretraining when labeled training data is limited.
arXiv Detail & Related papers (2022-08-05T15:06:26Z) - PatchNR: Learning from Small Data by Patch Normalizing Flow
Regularization [57.37911115888587]
We introduce a regularizer for the variational modeling of inverse problems in imaging based on normalizing flows.
Our regularizer, called patchNR, involves a normalizing flow learned on patches of very few images.
arXiv Detail & Related papers (2022-05-24T12:14:26Z) - CXR-FL: Deep Learning-based Chest X-ray Image Analysis Using Federated
Learning [0.0]
We present an evaluation of deep learning-based models for chest X-ray image analysis using the federated learning method.
We show that classification models perform worse if trained on a region of interest reduced to segmentation of the lung compared to the full image.
arXiv Detail & Related papers (2022-04-11T15:47:54Z) - Many-to-One Distribution Learning and K-Nearest Neighbor Smoothing for
Thoracic Disease Identification [83.6017225363714]
deep learning has become the most powerful computer-aided diagnosis technology for improving disease identification performance.
For chest X-ray imaging, annotating large-scale data requires professional domain knowledge and is time-consuming.
In this paper, we propose many-to-one distribution learning (MODL) and K-nearest neighbor smoothing (KNNS) methods to improve a single model's disease identification performance.
arXiv Detail & Related papers (2021-02-26T02:29:30Z) - Improved Slice-wise Tumour Detection in Brain MRIs by Computing
Dissimilarities between Latent Representations [68.8204255655161]
Anomaly detection for Magnetic Resonance Images (MRIs) can be solved with unsupervised methods.
We have proposed a slice-wise semi-supervised method for tumour detection based on the computation of a dissimilarity function in the latent space of a Variational AutoEncoder.
We show that by training the models on higher resolution images and by improving the quality of the reconstructions, we obtain results which are comparable with different baselines.
arXiv Detail & Related papers (2020-07-24T14:02:09Z)
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.