Ilia Buralkin is a PhD student in the Liu Lab at Baylor College of Medicine and Texas Children’s Hospital, specializing in deep learning for computer vision with applications in genomics and medical imaging.
His research applies computer vision and multimodal deep learning methods across three interconnected domains. In single-cell genomics, he develops scDeepVariant, a deep learning framework for germline variant calling in single-cell RNA sequencing data, addressing the challenges of detecting genetic variants from sparse and noisy scRNA-seq reads. In structural variant analysis, he builds multimodal machine learning pipelines for copy number variant (CNV) pathogenicity prediction, integrating image-encoded genomic features with clinical and phenotypic data for variant prioritization. In point-of-care ultrasound (POCUS), he develops computer vision systems for lung pathology prediction, applying video-based deep learning to clinical ultrasound interpretation using Texas Children’s Hospital operational data in weakly labeled scenarios.
Across these projects, Ilia’s work is unified by a focus on adapting computer vision and multimodal architectures to diverse biomedical data. His research emphasizes rigorous evaluation, methodological robustness, and bridging computational methods with clinical and biological applications.