What is Wrong with Continual Learning in Medical Image Segmentation?
- URL: http://arxiv.org/abs/2010.11008v1
- Date: Wed, 21 Oct 2020 13:48:37 GMT
- Title: What is Wrong with Continual Learning in Medical Image Segmentation?
- Authors: Camila Gonzalez, Georgios Sakas and Anirban Mukhopadhyay
- Abstract summary: Continual learning protocols are attracting increasing attention from the medical imaging community.
In a continual setup, data from different sources arrives sequentially and each batch is only available for a limited period.
We show that the benchmark outperforms two popular continual learning methods for the task of T2-weighted MR prostate segmentation.
- Score: 1.2020488155038649
- License: http://creativecommons.org/licenses/by-sa/4.0/
- Abstract: Continual learning protocols are attracting increasing attention from the
medical imaging community. In a continual setup, data from different sources
arrives sequentially and each batch is only available for a limited period.
Given the inherent privacy risks associated with medical data, this setup
reflects the reality of deployment for deep learning diagnostic radiology
systems. Many techniques exist to learn continuously for classification tasks,
and several have been adapted to semantic segmentation. Yet most have at least
one of the following flaws: a) they rely too heavily on domain identity
information during inference, or b) data as seen in early training stages does
not profit from training with later data. In this work, we propose an
evaluation framework that addresses both concerns, and introduce a fair
multi-model benchmark. We show that the benchmark outperforms two popular
continual learning methods for the task of T2-weighted MR prostate
segmentation.
Related papers
- Segmentation of Multiple Sclerosis Lesions across Hospitals: Learn
Continually or Train from Scratch? [8.691839346510116]
Experience replay is a well-known continual learning method.
We show that replay is able to achieve positive backward transfer and reduce catastrophic forgetting.
Our experiments show that replay is able to achieve positive backward transfer and reduce catastrophic forgetting.
arXiv Detail & Related papers (2022-10-27T00:32:13Z) - PCA: Semi-supervised Segmentation with Patch Confidence Adversarial
Training [52.895952593202054]
We propose a new semi-supervised adversarial method called Patch Confidence Adrial Training (PCA) for medical image segmentation.
PCA learns the pixel structure and context information in each patch to get enough gradient feedback, which aids the discriminator in convergent to an optimal state.
Our method outperforms the state-of-the-art semi-supervised methods, which demonstrates its effectiveness for medical image segmentation.
arXiv Detail & Related papers (2022-07-24T07:45:47Z) - LifeLonger: A Benchmark for Continual Disease Classification [59.13735398630546]
We introduce LifeLonger, a benchmark for continual disease classification on the MedMNIST collection.
Task and class incremental learning of diseases address the issue of classifying new samples without re-training the models from scratch.
Cross-domain incremental learning addresses the issue of dealing with datasets originating from different institutions while retaining the previously obtained knowledge.
arXiv Detail & Related papers (2022-04-12T12:25:05Z) - When Accuracy Meets Privacy: Two-Stage Federated Transfer Learning
Framework in Classification of Medical Images on Limited Data: A COVID-19
Case Study [77.34726150561087]
COVID-19 pandemic has spread rapidly and caused a shortage of global medical resources.
CNN has been widely utilized and verified in analyzing medical images.
arXiv Detail & Related papers (2022-03-24T02:09:41Z) - Federated Cycling (FedCy): Semi-supervised Federated Learning of
Surgical Phases [57.90226879210227]
FedCy is a semi-supervised learning (FSSL) method that combines FL and self-supervised learning to exploit a decentralized dataset of both labeled and unlabeled videos.
We demonstrate significant performance gains over state-of-the-art FSSL methods on the task of automatic recognition of surgical phases.
arXiv Detail & Related papers (2022-03-14T17:44:53Z) - Practical Challenges in Differentially-Private Federated Survival
Analysis of Medical Data [57.19441629270029]
In this paper, we take advantage of the inherent properties of neural networks to federate the process of training of survival analysis models.
In the realistic setting of small medical datasets and only a few data centers, this noise makes it harder for the models to converge.
We propose DPFed-post which adds a post-processing stage to the private federated learning scheme.
arXiv Detail & Related papers (2022-02-08T10:03:24Z) - Continual learning of longitudinal health records [0.0]
We evaluate a variety of continual learning methods on longitudinal ICU data.
We find that while several methods mitigate short-term forgetting, domain shift remains a challenging problem over large series of tasks.
arXiv Detail & Related papers (2021-12-22T15:08:45Z) - Adversarial Continual Learning for Multi-Domain Hippocampal Segmentation [0.46023882211671957]
Deep learning for medical imaging suffers from temporal and privacy-related restrictions on data availability.
We propose an architecture that leverages the simultaneous availability of two or more datasets to learn a disentanglement between the content and domain.
We showcase that our method reduces catastrophic forgetting and outperforms state-of-the-art continual learning methods.
arXiv Detail & Related papers (2021-07-19T10:55:21Z) - Towards Cross-modality Medical Image Segmentation with Online Mutual
Knowledge Distillation [71.89867233426597]
In this paper, we aim to exploit the prior knowledge learned from one modality to improve the segmentation performance on another modality.
We propose a novel Mutual Knowledge Distillation scheme to thoroughly exploit the modality-shared knowledge.
Experimental results on the public multi-class cardiac segmentation data, i.e., MMWHS 2017, show that our method achieves large improvements on CT segmentation.
arXiv Detail & Related papers (2020-10-04T10:25:13Z) - Semi-supervised Medical Image Segmentation through Dual-task Consistency [18.18484640332254]
We propose a novel dual-task deep network that jointly predicts a pixel-wise segmentation map and a geometry-aware level set representation of the target.
Our method can largely improve the performance by incorporating the unlabeled data.
Our framework outperforms the state-of-the-art semi-supervised medical image segmentation methods.
arXiv Detail & Related papers (2020-09-09T17:49:21Z) - Disentangling Human Error from the Ground Truth in Segmentation of
Medical Images [12.009437407687987]
We present a method for jointly learning, from purely noisy observations alone, the reliability of individual annotators and the true segmentation label distributions.
We demonstrate the utility of the method on three public medical imaging segmentation datasets with simulated (when necessary) and real diverse annotations.
arXiv Detail & Related papers (2020-07-31T11:03:12Z)
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.