One-Vote Veto: A Self-Training Strategy for Low-Shot Learning of a Task-Invariant Embedding to Diagnose Glaucoma

12/09/2020
by   Rui Fan, et al.
0

Convolutional neural networks (CNNs) are a promising technique for automated glaucoma diagnosis from images of the fundus, and these images are routinely acquired as part of an ophthalmic exam. Nevertheless, CNNs typically require a large amount of well-labeled data for training, which may not be available in many biomedical image classification applications, especially when diseases are rare and where labeling by experts is costly. This paper makes two contributions to address this issue: (1) It introduces a new network architecture and training method for low-shot learning when labeled data are limited and imbalanced, and (2) it introduces a new semi-supervised learning strategy that uses additional unlabeled training data to achieve great accuracy. Our multi-task twin neural network (MTTNN) can use any backbone CNN, and we demonstrate with ResNet-50 and MobileNet-v2 that its accuracy with limited training data approaches the accuracy of a finetuned backbone trained with a dataset that is 50 times larger. We also introduce One-Vote Veto (OVV) self-training, a semi-supervised learning strategy, that is designed specifically for MTTNNs. By taking both self-predictions and contrastive-predictions of the unlabeled training data into account, OVV self-training provides additional pseudo labels for finetuning a pretrained MTTNN. Using a large dataset with more than 50,000 fundus images acquired over 25 years, extensive experimental results demonstrate the effectiveness of low-shot learning with MTTNN and semi-supervised learning with OVV. Three additional, smaller clinical datasets of fundus images acquired under different conditions (cameras, instruments, locations, populations), are used to demonstrate generalizability of the methods. Source code and pretrained models will be publicly available upon publication.

READ FULL TEXT

page 12

page 13

research
06/03/2019

Learning to Self-Train for Semi-Supervised Few-Shot Classification

Few-shot classification (FSC) is challenging due to the scarcity of labe...
research
06/01/2023

Doubly Robust Self-Training

Self-training is an important technique for solving semi-supervised lear...
research
06/27/2020

PCLNet: A Practical Way for Unsupervised Deep PolSAR Representations and Few-Shot Classification

Deep learning and convolutional neural networks (CNNs) have made progres...
research
05/19/2020

Self-Updating Models with Error Remediation

Many environments currently employ machine learning models for data proc...
research
01/23/2020

Continual Local Replacement for Few-shot Image Recognition

The goal of few-shot learning is to learn a model that can recognize nov...
research
03/18/2020

Task-Adaptive Clustering for Semi-Supervised Few-Shot Classification

Few-shot learning aims to handle previously unseen tasks using only a sm...
research
04/14/2023

SMAE: Few-shot Learning for HDR Deghosting with Saturation-Aware Masked Autoencoders

Generating a high-quality High Dynamic Range (HDR) image from dynamic sc...

Please sign up or login with your details

Forgot password? Click here to reset