
Ranit Karmakar
@iranitk
Specialist Postdoc @harvardmed👨🏽‍🎓
Medical/Bio-image AI Researcher 👨🏾‍💻
Health Tech Founder 🤵🏾‍♂️
ID: 1524030300547952640
https://iac.hms.harvard.edu/ 10-05-2022 14:16:02
12 Tweet
19 Followers
74 Following

I2K (i2kconference.org) group picture time!!! Look at this AMAZING crowd!!! 🤩 Human Technopole #I2K2024


Eager to learn how to train your own deep learning models for image segmentation? We got you covered! We are organizing a new workshop "Build Your Own Image Segmentation Pipeline in Python" on December 9th 2024 in-person Harvard Medical School. Apply here: iac.hms.harvard.edu/teaching/2024-…



Announcing the Boston Bioimage Analysis Course — BoBiAC! Join us this July at Harvard Medical School in Boston for a 6-day deep dive into bioimage analysis with Python, hosted by HMS Image Analysis Collaboratory and CITE. Apply by May 16th 2025!


‼BREAKING‼ New funding secured for the Boston Bioimage Analysis Course (BoBiAC) this July at Harvard Medical School. Fees down 37%, with waivers for academic applicants, for an intensive hands-on intro to bioimage analysis with Python! Apply by June 1st! iac.hms.harvard.edu/bobiac/2025


💻 Apply to our beginner-friendly Boston Bioimage Analysis Course! Get started with Python and learn to analyze fluorescence microscopy images—no coding experience required! Fee waivers available thanks to funding from BioImaging North America! 🗓️ Apply June 1st! iac.hms.harvard.edu/bobiac/2025/


🚨 New paper out: SoftPQ – a robust metric for instance segmentation 🧠📊 Hard IoU thresholds miss the nuance. We propose: ✔️ Soft matching ✔️ Tunable thresholds ✔️ Sublinear penalties Rovust, interpretable evaluation. 📄 Paper: ⬇️ arxiv.org/abs/2505.12155 w/ Simon F. Nørrelykke


✅ Quality control of segmentation & ✨ Feature extraction (intensity, morphology, spatial features) by Antoine A. Ruzette & Ranit Karmakar 🔧 Workflow building with ZeroCostDL & Bioimage Zoo 📊 Data management strategies from Nick Condon (IMB, Australia)
