Embracing Imperfect Datasets: A Review of Deep Learning Solutions for Medical Image Segmentation

TL;DR

This review discusses recent deep learning methods for medical image segmentation that effectively handle imperfect datasets with limited or noisy annotations, highlighting technical innovations and empirical results.

Contribution

It provides a comprehensive overview of solutions for training segmentation models on imperfect datasets, comparing their benefits and requirements.

Findings

Various techniques improve segmentation with scarce annotations

Methods effectively handle noisy and weak labels

Survey highlights best practices and future directions

Abstract

The medical imaging literature has witnessed remarkable progress in high-performing segmentation models based on convolutional neural networks. Despite the new performance highs, the recent advanced segmentation models still require large, representative, and high quality annotated datasets. However, rarely do we have a perfect training dataset, particularly in the field of medical imaging, where data and annotations are both expensive to acquire. Recently, a large body of research has studied the problem of medical image segmentation with imperfect datasets, tackling two major dataset limitations: scarce annotations where only limited annotated data is available for training, and weak annotations where the training data has only sparse annotations, noisy annotations, or image-level annotations. In this article, we provide a detailed review of the solutions above, summarizing both the technical novelties and empirical results. We further compare the benefits and requirements of the surveyed methodologies and provide our recommended solutions. We hope this survey article increases the community awareness of the techniques that are available to handle imperfect medical image segmentation datasets.