Skip to content
View JKaestelHansen's full-sized avatar

Block or report JKaestelHansen

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
JKaestelHansen/README.md

Hi there 👋

👋 Hi, I'm Jacob Kæstel-Hansen (Kaestel-Hansen

Postdoctoral Fellow @ MIT | Bridging AI, Physics, and Biology


🚀 About Me

I'm a postdoc fellow at MIT, passionate about pushing the boundaries between artificial intelligence, physics, and biology. My mission is to make machine learning not only powerful but also interpretable—especially for complex biological data with limited labels. I thrive on exploring explainable AI (XAI) and uncertainty estimation to unlock new insights from images, motion, and more.

  • 🔬 Research Focus: Explainable AI & interpretability for biological data
  • 🧠 Fun Project: DeepSPT — applying deep learning to analyze biological motion
  • 🧑‍💻 Always Coding: Python, PyTorch, Computer Vision, Machine Learning, Deep Learning

🛠️ Skills & Technologies

AI Deep Learning Computer Vision XAI Uncertainty Estimation Statistics Machine Learning Time-Series Bioinformatics Image Analysis Python PyTorch TensorFlow NumPy Captum Pandas Scikit-Learn Matplotlib Jupyter


📚 Featured Projects

  • DeepSPT
    Leveraging deep learning to analyze and interpret biological motion: SEMORE

  • XAI Timeseries Model
    Exploring explainable AI and interpretability for biological datasets with limited labels.

  • Probabilistic ML
    Probabilistic approaches to machine learning and uncertainty.

  • LatticeProcessing
    Advanced tools for lattice-based data analysis.

  • Computer_vision) Playground for Dino and other computervision tools.

  • protein_regression
    A systematic analysis of regression models for protein engineering Repository to replicate the results of: A systematic analysis of regression models for protein engineering

  • SEMORE
    A multi-independent-module pipeline for structure segmentation and disection in single molecule localization microscopy (SMLM) data and the extraction of unique morphological fingerprints: SEMORE._

  • Variations_of_VAE
    Implementations of VAEs based on Riesselman et al 2018 "Deepsequence" both the standard from the paper, but also extended to conditional and semi-supervised variants.

  • Video_object_tracking_
    Hacking into pytorch maskrcnn for making an end2end model for object tracking over time and segmentation of morphology.

  • Additional analysis repos Multiple codebases for published studies from my PhD/Postdoc are also stored on my Github.


🌱 Mission Statement

“Bridging disciplines to make AI smarter, more transparent, and useful for unraveling the mysteries of biology.”


🔗 Connect with Me

LinkedIn

Google Scholar


⚡ Fun Facts

  • 🥾 Hiking enthusiast
  • 📚 Avid reader
  • 🧗 Rock climber & boulderer

Pinned Loading

  1. DeepSPT DeepSPT Public

    Python 8 5

  2. protein_regression protein_regression Public

    Forked from MachineLearningLifeScience/protein_regression

    The codebase to replicate the analysis of "A systematic analysis of regression models for protein engineering" (2024).

    Jupyter Notebook

  3. hatzakislab/SEMORE hatzakislab/SEMORE Public

    SEMORE is a multi-layered density-based clustering module to dissect biological assemblies and a morphology fingerprinting module for quantification by multiple geometric and kinetics-based descrip…

    Python 5

  4. XAI_Timeseries_model XAI_Timeseries_model Public

    Explainable AI model (XAI) for timeseries data.

    Jupyter Notebook

  5. Probabilistic_ML Probabilistic_ML Public

    This repo explores probabilistic ML in the context of predictions of diffusional properties with a specific interest in predicting Cohesin extrusion speeds.

    Python