Kristin Gallik, Ph.D.
Machine Learning & Computer Vision Scientist
Developing AI-powered solutions for complex bioimage analysis
Hello and welcome! ​
I build custom deep learning models and analysis pipelines that transform microscopy data into quantitative biological insights. Specializing in instance segmentation, transfer learning, and multimodal bioimage datasets.
A Bit About Me
I'm a computational scientist who bridges the gap between experimental biology and machine learning. With a PhD in biological sciences and 4 years of applied experience in Python, deep learning, and computer vision, I bring a unique combination of domain expertise and technical skills to challenging bioimage analysis problems.
​
Currently: Bioinformatics Scientist (Bioimage Analysis Specialist) at Van Andel Institute, where I develop ML solutions for ~300 scientists across 40+ research labs.
​
Seeking: Opportunities to advance machine learning and computer vision approaches for bioimage and multimodal biological datasets.
What I do
D.L & C.V.
-
Custom instance and semantic segmentation models
-
Transfer learning with CellPose, StarDist, U-Net architectures
-
Image classification with knowledge extraction
-
Multi-modal data analysis
Pipeline Development
-
End-to-end Python-based analysis workflows
-
Digital pathology whole slide image analysis
-
3D volumetric and time-series microscopy data
-
Reproducible, scalable solutions
Domain Expertise
-
Experimental design consultation
-
Understanding imaging artifacts and limitations
-
Biological context for model development
-
Training and education for researchers
Featured Work
Creating the First Computational Model of Mammary Gland Involution
Developing novel ML approaches for 1,200+ whole slide images, including pixel classifiers, spatial modeling, and deep learning classification with knowledge extraction.
_edited.jpg)
3D Microglia Segmentation with Hybrid Deep Learning
Custom transfer learning approach combining 2D CellPose models with 3D stitching to overcome sparse z-sampling challenges.
Automated Brain Atlas Registration & Quantification
Leveraging BrainGlobe platform to register and quantify pathological markers across 200+ brain regions.
Technical Expertise
​Languages & Tools: Python (primary), ImageJ Macro, Groovy
ML Frameworks: CellPose, StarDist, scikit-image, pyclesperanto
Platforms: GitHub, JIRA, digital pathology tools
Imaging: Confocal, TEM, SPIM, digital pathology, widefield to super-resolution
Training: Graduate of DL@MBL (AI@MBL) 2023 - intensive deep learning course for bioimage analysis
Connect with me
GitHub: https://github.com/vaioic
LinkedIn: https://www.linkedin.com/in/kristingallik/
Email: kristin[dot]gallik[at]gmail[dot]com
​
I'm actively seeking opportunities in machine learning and computer vision
for biological imaging. Let's connect if you're working on challenging problems
at the intersection of biology, imaging, and AI.