I'm leading a new BraTS on post-treatment glioma

Let me know if you are interested in being an annotator

The American Society of Neuroradiology American Society of Functional Neuroradiology Spyridon Bakas Rachit Saluja Evan Calabrese, MD PhD Ujjwal Baid Mariam Aboian Philipp Vollmuth Nourel Hoda Tahon Andreas Rauschecker Niky Farid Cortechs.ai Aly Abayazeed

Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.

Our review article on AI in digital pathology is out in Nature Reviews Electrical Engineering:
nature.com/articles/s4428…

We have faculty openings in AI for medical imaging at Cornell University in NYC. Applications submitted this week will receive full consideration: higheredjobs.com/details.cfm?Jo…

Proud to announce:
Mixtral 8x7B -- Mixtral of Experts

- Free to use under Apache 2.0 license
- outperforms Llama 2 70B with 6x faster inference.
- matches or outperforms GPT3.5
- masters English, French, Italian, German and Spanish.
- seq_len = 32K

mistral.ai/news/mixtral-o…

1/N

I am on the academic job market for 2024!

My vision is to develop powerful AI models for healthcare that are safe, reliable, and trustworthy.

E.g., my recent paper: arxiv.org/abs/2310.01685

Would appreciate any RTs and pointers!

See below for an overview of my prior work:

No more manually sorting brain MRI sequences!?

github.com/neuroAI-HD/HD-…

Andreas Rauschecker Evan Calabrese, MD PhD Rachit Saluja Spyridon Bakas Ujjwal Baid Rachit Saluja

Worried your labeled training dataset isn't big enough?? Congrats to Stephen Wahlig, UCSF Imaging resident, on this paper showing us how to efficiently fine-tune a U-net trained for a different purpose to #MS lesion segmentation. UCSF Center for Intelligent Imaging Jeff Rudie, MD PhD frontiersin.org/articles/10.33…

📢Calling all brain tumor researchers! Join our global federated learning study for automated volumetric response assessment in post-operative glioblastoma led by Dr. Evan Calabrese (Evan Calabrese, MD PhD) here at #DukeRadiology .

Are you a PhD researcher in the field of computational medical imaging? Are you looking for faculty jobs? We, at Weill Cornell Medicine Radiology, are hiring. Please read on. 🧵1/

Happy to see this longitudinal post-treatment glioma dataset be released!
imagingdatasets.ucsf.edu/dataset/2

Tons of effort went in to create it
Brandon K.K. Fields, M.D. Evan Calabrese, MD PhD Andreas Rauschecker Christopher Hess John Mongan UCSF Center for Intelligent Imaging

2024 BraTS!
Spyridon Bakas Ujjwal Baid

Congrats to all of the winners of the 2023 BraTS!

Spyridon Bakas Ujjwal Baid Evan Calabrese, MD PhD Hongwei Bran Li Mariam Aboian Rachit Saluja Ahmed W. Moawad Anahita Fathi Kazerooni CAMERA MRI AFRICA Anastasia Janas

The American Society of Neuroradiology MICCAI
Cortechs.ai Neosoma Inc.
MLCommons Sage Bionetworks

Very excited to present the new clinically relevant lesion-wise performance metrics for the 5 segmentations tasks BraTS this year along with Jeff Rudie, MD PhD at #MICCAI2023 .
Ujjwal Baid, Spyridon Bakas

So you claim your ML for medical imaging (MLMI) model is interpretable... but what does that really mean? What are the core, underlying elements of interpretable MLMI?

A thread 👇

🧵1/6

Everyone agrees that interpretability or explainability is critical in ML for medical imaging (MLMI). In our brand new paper, we formalize and present the needs and elements of Interpretability in Machine Learning for Medical Imaging: arxiv.org/abs/2310.01685

I will probably onboard one PhD student for science of science and one to fit big models to neural data/ neuroAI. Relevant prior experience and love for nerding out/ building algorithms and math will be needed.

Presenting KeyMorph: a deep learning-based, multimodal image registration framework that relies on automatically detecting corresponding keypoints!

To appear in Medical Image Analysis. w/ Evan Yu, Adrian Dalca, and Mert R. Sabuncu 🤖🩻🟧

🧵1/7

It's been my pleasure helping organize this years RSNA AI Challenge on Abdominal Trauma
press.rsna.org/timssnet/media…

Errol Colak John Mongan Adam Flanders, MD Luciano M Prevedello MD MPH

There are already >180 teams competing on Kaggle!
kaggle.com/competitions/r…

Those confident predictions can be scary. Use FALCON to explain and debug :)
Aloha! I’ll be at Honolulu next week to present our #ICML2023 work titled “Identifying Interpretable Subspaces for Image Representations”. Stop by our poster at Exhibit Hall 1 #705 at 2pm on Tue 07/25!

The BraTS 2023 meningioma segmentation challenge is in full swing with over 1000 training cases and 141 validation cases now publicly available. Participate and make meningioma great: synapse.org/brats
BraTS #ForgottenBrainTumor #MeningiomaRespect