About
Esther Jin is a Research Assistant II at Cedars-Sinai working at the intersection of artificial intelligence, computational pathology, and biomedical data science. Her current research focuses on developing multimodal deep learning approaches that integrate histology imaging and spatial transcriptomics to better understand spatial patterns of gene expression in tissue. She has experience developing end-to-end PyTorch pipelines, training computer vision models, and applying pathology foundation models for image representation learning and spatial prediction. Previously, she conducted research in self-supervised learning for computational pathology, where she developed a DINO-based model to learn representations from thousands of unlabeled cytology images across multiple tumor types. Her work has included image embedding analysis, clustering, multimodal data integration, and spatial visualization. She has also contributed to public health and healthcare data projects involving epidemiological datasets, provider enrollment, and data quality analysis. She is currently pursuing a Master of Science in Data Science in Biomedicine at the University of California, Los Angeles. She is interested in biomedical artificial intelligence, computational biology, medical imaging, multimodal learning, and applications of machine learning to improve biomedical research and clinical understanding.