Hemepath & Clinical Informatics Fellow @Nu_Pathology | Passionate about Digital Pathology & AI | Merging medicine & technology for transformative patient care
AI in pathology is evolving! From general-purpose models to hypothesis-driven approaches, we’re unlocking new ways to understand disease.
Check out the full plenary session from Dr. Lee Cooper on our website—coming soon! #PathVisions24
#PathMatch2025 tomorrow is the day! All programs will use #Thalamus! For those who receive an invitation from @NU_Pathology - please ensure that you schedule yourself for an interview day as well as for the Resident Meet & Greet to be held on the same week!
Based on numerous requests, we are providing the open ShareIT link for UNI and CONCH. Please access it below:
Open ShareIT Read Links:
UNI: https://t.co/lE3cYSUJoY
CONCH: https://t.co/fvA1jNe0om
Journal Links for complete pdf:
UNI: https://t.co/9eHXwk0kjJ
CONCH: https://t.co/f207RP1hA0
It's amazing how much information can be gleaned via #AI from pathology slides (WSI), as reviewed @AI4Pathology and colleagues, including prognosis, predicting response to therapy, and major molecular features of a patient's cancer
https://t.co/JAmf5M26tN
Good News from Chicago!! Work by @NU_Pathology members Mohamed Tageldin, Lee Cooper and team on predictive modeling of breast cancer outcomes using AI and pathology data makes front page of @chicagotribune ! Talent + Care + Commitment = Advances in Patient Care.
Interested in hearing from our current residents? Join us for a virtual Q&A on 9/6/23 at 5pm CST. Submit your questions ahead of time and register here: https://t.co/Jm89HP4Laj
#PurplePath@Path_SIG@McGawGME
⚡️Excited to share our new @NatureMedicine paper where we used Twitter to build a vision-language foundation #AI for #pathology https://t.co/1DQJ1zzDwh
We curated >100K public Twitter threads w/ medical images+text to create PLIP for semantic search and 0-shot pred.
All our data (OpenPath): https://t.co/J5x8gxx5QC
Open source PLIP: https://t.co/feIxSan7nv
Getting sufficient data is a key bottleneck for medical AI. This work shows how we can harness the tremendous medical knowledge openly shared on social media to develop models. We used multiple filters to improve Twitter data quality. PLIP performs well for 0-shot classifications, fine-tuning, and image and text-based retrievals.
Great work by @ZhiHuangPhD@federicobianchy@mertyuksekgonul and Tom Montine 👏 And thanks to support from @StanfordHAI@StanfordMed!
Check out the #AI#deeplearning powered Immunohistochemistry Nuclei Counting App now on CODIDO (https://t.co/tVaGhofU82). Simply upload an extracted image from a #digitalpathology slide and get a objective quantification of positive nuclei. #pathology#pathtwitter#neuropath
Interested in a pathology residency at Northwestern Pathology? Join Chairman @DanielJBrat and Residency Program Leadership for a Q&A on 8/24/23 from 10-11am CST! Register here: https://t.co/VYOxvvUgM0…
#PurplePath#PathMatch24#PathTwitter#Path2Path 💜
Excited to announce CONCH, a new visual language foundation model for #pathology, trained with 1.17 Million pathology image / caption pairs and achieves SOTA performance on zero-shot classification, text-to-img retrieval, segmentation and more!
Pre-print: https://t.co/N1VsA0OiYT
Thrilled to share our new @natBME review article discussing algorithmic #fairness in #AI for medicine and healthcare. We discuss sources of algorithmic biases in healthcare, and emerging methods for mitigating biases. Led by our superstar grad student @richardjchen
Journal Link: https://t.co/xCQn03xGdB
Our newest product, Paige Colon MSI, will support pathologists in the detection of MSI status in colon cancer samples using H&E alone. https://t.co/eYiY5fw7Bm