I am passionate about bridging the gap between biology and code. I am really interested in extracting quantitative insights from complex microscopy data and making sense of it in a data-driven manner.
My Core Focus: Image & Data Analytics
- Bioimage Analysis: My focus is on extracting biologically useful quantitatve data in a reproducible and accurate manner! Datasets can be of high-dynamic-range, multi-dimensional (3D/4D/5D) data where signal-to-noise ratios are low and "ground truth" is hard to define. They can also be quite big, in the order of 100s of GB to TBs!
- Data Analytics: Once data is extracted from images, I really enjoy trying to making sense of it. Its often a feedback loop, as it informs the image analysis! My current technical interests lie in Deep Learning, unsupervised representation learning, dimensionality reduction techniques (UMAP, PHATE!).
- 🔭 Current Focus: Developing automated image analysis pipelines and exploring unsupervised learning strategies for phenotypic profiling.
- 🔧 I Build: User-friendly, open-source tools that democratize advanced image analysis for biologists.
- 👯 Collaborate: Always interested in projects involving extracting and analysing multidimensional data.
- 🤔 Interest: Identifying robust methods for unsupervised representation learning and analyzing latent spaces in multivariate data.
- 💬 Ask me about: The difference between "pretty images" and "quantifiable data," image segmentation, and reproducible research.
- Computer Vision: Segmentation, Feature Extraction, Object Tracking, Registration, Image Restoration (Denoising/Deconvolution).
- Data Science: Deep Learning, Dimensionality Reduction, Multivariate Analysis, Hypothesis Testing, Visualization,.
- Stack: Python (NumPy, Pandas, Scikit-learn, PyTorch), ImageJ/Fiji, Napari, QuPath.