Cross comparison representation learning for semi-supervised segmentation of cellular nuclei in immunofluorescence staining

The morphological analysis of cells from optical images is vital for interpreting brain function in disease states. Extracting comprehensive cell morphology from intricate backgrounds, common in neural and some medical images, poses a significant challenge. Due to the huge workload of manual recognition, automated neuron cell segmentation using deep learning algorithms with labeled data is integral to neural image analysis tools. To combat the high cost of acquiring labeled data, we propose a novel semi-supervised cell segmentation algorithm for immunofluorescence-stained cell image datasets (ISC), utilizing a mean-teacher semi-supervised learning framework. We include a “cross comparison representation learning block” to enhance the teacher–student model comparison on high-dimensional channels, thereby improving feature compactness and separability, which results in the extraction of higher-dimensional features from unlabeled data. We also suggest a new network, the Multi Pooling Layer Attention Dense Network (MPAD-Net), serving as the backbone of the student model to augment segmentation accuracy. Evaluations on the immunofluorescence staining datasets and the public CRAG dataset illustrate our method surpasses other top semi-supervised learning methods, achieving average Jaccard, Dice and Normalized Surface Dice (NSD) indicators of 83.22%, 90.95% and 81.90% with only 20% labeled data. The datasets and code are available on the website at https://github.com/Brainsmatics/CCRL. •We proposed a semi supervised learning framework based on cross comparison representation learning for the first time. This framework enhances the segmentation ability of the network by comparing learning algorithms to enhance the feature fitting before the teacher-student model in a semi-supervised learning framework.•We also provided a public ISC dataset for facilitating the semi-supervised cell segmentation research, which contains 2550 immunofluorescence staining images and human-labeled ground truth.•We tested the proposed method on the ISC and CRAG datasets. It achieves average Jaccard, Dice and Normalized Surface Dice (NSD) indicators of 83.22%, 90.95% and 81.90% with only 20% labeled data on ISC and CRAG datasets, which surpasses other state-of-the-art semi-supervised learning methods.