Biologically-inspired Semi-supervised Semantic Segmentation for Biomedical Imaging

TL;DR

This paper introduces a bio-inspired semi-supervised learning method for biomedical image segmentation that combines Hebbian unsupervised feature discovery with supervised fine-tuning, outperforming state-of-the-art methods especially with limited labeled data.

Contribution

It presents a novel two-stage training approach using Hebbian learning for unsupervised feature extraction followed by supervised fine-tuning, improving segmentation performance in biomedical imaging.

Findings

Outperforms SOTA methods across various label levels

Unsupervised stage improves initialization for supervised training

Effective in data-scarce biomedical imaging scenarios

Abstract

We propose a novel bio-inspired semi-supervised learning approach for training downsampling-upsampling semantic segmentation architectures. The first stage does not use backpropagation. Rather, it exploits the Hebbian principle ``fire together, wire together'' as a local learning rule for updating the weights of both convolutional and transpose-convolutional layers, allowing unsupervised discovery of data features. In the second stage, the model is fine-tuned with standard backpropagation on a small subset of labeled data. We evaluate our methodology through experiments conducted on several widely used biomedical datasets, deeming that this domain is paramount in computer vision and is notably impacted by data scarcity. Results show that our proposed method outperforms SOTA approaches across different levels of label availability. Furthermore, we show that using our unsupervised stage to initialize the SOTA approaches leads to performance improvements. The code to replicate our experiments can be found at https://github.com/ciampluca/hebbian-bootstraping-semi-supervised-medical-imaging