My PhD research investigates high-risk neuroblastoma (NB) and malignant paraganglioma (PGL), both of which originate from the sympatho-adrenal lineage. Neuroblastoma is a pediatric malignancy characterized by marked clinical heterogeneity, ranging from spontaneous regression to aggressive metastatic disease. Paraganglioma, by contrast, is a rare neuroendocrine tumor with distinct biological features. In our laboratory, we have collected human tissue samples from both NB and PGL patients.
To analyze these tumors at single-cell resolution while preserving spatial context, we apply an integrated approach combining single-cell RNA sequencing and spatial transcriptomics. This strategy enables us to dissect the tumor microenvironment, spatial cellular organization, clonal architecture, and cell–cell interactions.
A central component of this work involves the use of artificial intelligence (AI) and machine learning (ML) techniques for processing and interpreting the high-resolution spatial transcriptomic images. Specifically, we employ deep learning models—such as convolutional neural networks (CNNs)—for accurate cell segmentation, enabling the precise identification of individual cells within complex tissue sections. Furthermore, we apply ML-based clustering and spatial pattern recognition algorithms to identify cellular niches—regions of the tissue where specific cell types co-localize or interact in functionally significant ways. These AI/ML tools are essential for extracting biologically meaningful insights from large-scale spatial data that would be infeasible to analyze manually.