Imbalance-Aware Self-Supervised Learning for 3D Radiomic Representations

03/06/2021
by   Hongwei Li, et al.
19

Radiomic representations can quantify properties of regions of interest in medical image data. Classically, they account for pre-defined statistics of shape, texture, and other low-level image features. Alternatively, deep learning-based representations are derived from supervised learning but require expensive annotations from experts and often suffer from overfitting and data imbalance issues. In this work, we address the challenge of learning representations of 3D medical images for an effective quantification under data imbalance. We propose a self-supervised representation learning framework to learn high-level features of 3D volumes as a complement to existing radiomics features. Specifically, we demonstrate how to learn image representations in a self-supervised fashion using a 3D Siamese network. More importantly, we deal with data imbalance by exploiting two unsupervised strategies: a) sample re-weighting, and b) balancing the composition of training batches. When combining our learned self-supervised feature with traditional radiomics, we show significant improvement in brain tumor classification and lung cancer staging tasks covering MRI and CT imaging modalities.

READ FULL TEXT

page 3

page 8

research
10/05/2019

Self-supervised Feature Learning for 3D Medical Images by Playing a Rubik's Cube

Witnessed the development of deep learning, increasing number of studies...
research
01/30/2023

Exploring Image Augmentations for Siamese Representation Learning with Chest X-Rays

Image augmentations are quintessential for effective visual representati...
research
01/04/2022

Self-supervised Learning from 100 Million Medical Images

Building accurate and robust artificial intelligence systems for medical...
research
03/28/2023

Data Efficient Contrastive Learning in Histopatholgy using Active Sampling

Deep Learning based diagnostics systems can provide accurate and robust ...
research
02/28/2020

Self-supervised Representation Learning for Ultrasound Video

Recent advances in deep learning have achieved promising performance for...
research
12/14/2020

Aggregative Self-Supervised Feature Learning

Self-supervised learning (SSL) is an efficient approach that addresses t...
research
11/08/2021

Hybrid BYOL-ViT: Efficient approach to deal with small datasets

Supervised learning can learn large representational spaces, which are c...

Please sign up or login with your details

Forgot password? Click here to reset