Enhancing Cell Segmentation using Deep Learning Models by Custom Processing Techniques

This study presents a comprehensive workflow for cell segmentation using the Monuseg dataset. Our method integrates several advanced techniques to enhance the performance of segmentation models. We apply Principal Component Analysis (PCA) and Gaussian Mixture Model - Expectation Maximization (GMM-EM) clustering for data pre-processing. The training and evaluation significantly improved the mean IoU metric, demonstrating the enhanced performance of Language meets Vision Transformer (LViT) compared to other deep-learning networks. This constructed workflow shows substantial potential in improving the accuracy of cell segmentation tasks, which is crucial and beneficial for various biomedical applications.