A comparative study of semi- and self-supervised semantic segmentation of biomedical microscopy data
This work addresses the challenge of scarce labeled data in biomedical microscopy, offering a comparative analysis of alternative training methods, though it appears incremental as it adapts existing techniques.
The study tackled the problem of training CNNs for biomedical image segmentation with limited labeled data by adapting semi- and self-supervised methods, resulting in validated performance improvements for semantic segmentation on microscopy images.
In recent years, Convolutional Neural Networks (CNNs) have become the state-of-the-art method for biomedical image analysis. However, these networks are usually trained in a supervised manner, requiring large amounts of labelled training data. These labelled data sets are often difficult to acquire in the biomedical domain. In this work, we validate alternative ways to train CNNs with fewer labels for biomedical image segmentation using. We adapt two semi- and self-supervised image classification methods and analyse their performance for semantic segmentation of biomedical microscopy images.