Welcome
As a research scientist currently engaged in postdoctoral research at the Minnesota Institute for Astrophysics, University of Minnesota (UMN), I collaborate with the Zooniverse citizen-science platform team. My work spans diverse fields, including astronomy, high-energy astrophysics, deep learning, and biomedicine. I specialize in creating innovative machine learning approaches and citizen-science methods for identifying anomalous and scientifically significant subjects in large datasets.
In astronomy, my research explores the birth, growth, and transformation of galaxies, utilizing image processing techniques and novel deep learning methods. My Ph.D. work involved quantifying the collision rate of galaxies in the Universe using data from the Hubble Space Telescope.
​
In the realm of citizen science, I focus on accelerating the scientific output of various crowd-sourcing projects across disciplines through "human-in-the-loop" machine learning methods. Specifically, my research aims to expedite the discovery of unknown, scientifically interesting phenomena in data.
​
I am experienced in the vast arena of AI & deep learning, spanning supervised & unsupervised methods applied to various image-based computer vision tasks, generative and sequential data modeling using GANs, Vision Transformers. I am keen on developing new frameworks to help solve real-world challenges.
​
Simultaneously, I am deeply passionate about biomedical topics, particularly pathology and diagnostic techniques. My work involves applying deep learning techniques to automate segmentation of targets in biomedical imaging data (e.g., microscopy, MRI), such as cellular organelles and tumors. Additionally, I contribute to research on the pathogenesis of diseases like viral hepatitis and endometriosis.