Derive useful insights with our cutting-edge techniques
We use our own developed deep learning techniques for high performance single cell data visualization, clustering and cell annotation. We have included classical methods such PCa, t-SNE, PHATE and UMAP for performance comparison.
See exampleWe have developed techniques for analyzing the high dimensional feature space data from neural networks in classification, regression, segmentation, superresolution and prediction tasks. These techniques improve the interpretability and performance of the black box neural networks.
See exampleWe have developed deep learning techniques for accurate analysis of tabular data in both classification and regression tasks. Our methods convert the tabular data into images and apply 2D CNN for high performance and interpretable analysis.
See example