Favour Nerrise 🇨🇲 (@fnerrise) header
Favour Nerrise 🇨🇲 (@fnerrise) profile picture

@Stanford_EE PhD Candidate | Digital Biomarkers x AI x NeuroTech Digital Health | Prev. Amazon, Meta, NASA JPL

Joined April 2012
Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)Photo by Favour Nerrise 🇨🇲 (@fnerrise)
There’s nothing like a morning in SF! ❌ A man yelled that I’m sick in the head. ✅ A bus driver offered me a banana!
1
11
451
Favour Nerrise 🇨🇲 retweeted
Stanford Digital Brain Project has launched! We're hoping to make a seriously useful model of the human brain. Please come join us! news.stanford.edu/stories/20…
20
123
20
712
72,213
Favour Nerrise 🇨🇲 retweeted
1/ Can AI help doctors make more informed decisions based on patient activity and behavior over the course of a hospital stay? Today we introduce TRACER: a training-free multimodal harness that enables LLMs and VLMs to track patient activity using bedside video and health records. Blogpost: ai.stanford.edu/blog/tracer/ Paper: arxiv.org/abs/2610.21266
10
38
9
190
17,997
Favour Nerrise 🇨🇲 retweeted
LLMs have tokens. Images have patches. What should the brain have? 🤔 HBAR, our hierarchical tokenizer for fMRI, was just accepted to NeurIPS 2026! 🎉 It compresses fMRI ~415× while preserving structure from voxels → parcels → functional networks.
1
5
13
498
🙌🏿 Looks I picked the right dissertation topic. Loving all these Google Research Health contributions.
Finding clinical signals in large-scale health sensing data streams is quite challenging. We introduce a biomarker discovery framework that pairs hypothesis generation with adversarial validation to discover rigorous digital biomarkers alongside human experts. Read more: goo.gle/4gavroG
1
11
533
Favour Nerrise 🇨🇲 retweeted
Curriculum learning changes the data. Progressive stacking changes the model. Human development changes both at once. What happens when language model pretraining does the same? We explore this in Curriculum-Guided Layer Scaling (CGLS) 🧵
5
17
1
59
7,131
Favour Nerrise 🇨🇲 retweeted
The QuantiPhy Challenge is now officially open! Register your team at: quantiphy.stanford.edu/compe… Submit your results to win amazing prizes!!!
Thrilled to announce the 🏎️QuantiPhy Challenge🏃 to appear in #NeurIPS2026 Competition Track! Read more at: quantiphy.stanford.edu/compe… We host this competition to bring together attention from two important communities: VLMs and physics reasoning. How numerically accurate can your model estimate size/velocity/acceleration from video? Submit your results to the Quantiphy Challenge!! P.S. We are seeking sponsors to prepare exciting prizes for winners! Please reach out for sponsors that are open to collaborate :)
6
1
13
3,714
Favour Nerrise 🇨🇲 retweeted
I am thrilled to share that UC Berkeley and UCSF have launched a joint initiative in Computational Biomedicine! cdss.berkeley.edu/news/uc-be… We will soon be recruiting new faculty and postdoctoral fellows. Please repost to help spread the word.
2
56
2
239
37,504
Favour Nerrise 🇨🇲 retweeted
Thrilled to share that our paper on AMIE—our conversational medical AI—is now published in @Nature. We wanted to tackle one of the most challenging areas in clinical AI: management reasoning over multiple patient visits. nature.com/articles/s41586-0…
9
56
2
202
22,568
Favour Nerrise 🇨🇲 retweeted
🚨 New in Nature Medicine 🚨 Difficulty walking is one of the most disabling symptoms of Parkinson's disease—and one of the hardest to treat. Today we're excited to share our new study showing that deep brain stimulation (DBS) can be synchronized to each step a person takes.
7
25
1
75
5,974
Favour Nerrise 🇨🇲 retweeted
The AI Startups in Africa dataset has been accepted to the Deep Learning Indaba 2026 Dataset Showcase! @DeepIndaba This is the first conference the dataset will be formally presented at, and I anticipate great conversations. doi.org/10.7910/DVN/Z0PKGO
2
11
47
1,719
Excited to be presenting our #CVPR2026 CV4Clinic workshop paper TODAY!✨ 🧠GeoSAE: Geometric Prior-Guided Layer-Wise Sparse Autoencoder Annotation of Brain MRI Foundation Models 📍Poster 284 at 2:40 PM in Exhibit Hall A 📃: arxiv.org/abs/2605.01829 💻: github.com/favour-nerrise/Ge…
1
3
14
553
6/ ⚠️Another problem is age confounding in Alzheimer’s, where many features look disease-related even when they’re not. ✅ GeoSAE annotates features using age-adjusted clinical associations, & checks whether features relate to AD, sex, genetics, comorbidities, aging, etc.
1
2
47
7/ Across ~13k 3D brain MRI scans from ADNI and AIBL, GeoSAE finds a compact set of interpretable features that predicts MCI-to-AD conversion and replicates across cohorts. 🫂 Huge thanks to my amazing collaborators Lucy Yin, Mohammad H. Abbasi, Kilian M. Pohl, and @eadeli!
1
4
296
2/ Brain MRI foundation models can learn rich representations of anatomy, but it's still hard to know what clinical signals they are actually using. For Alzheimer’s disease, this is especially important because age, brain structure, and disease progression are deeply linked.
1
18
3/ In 🧠GeoSAE, we use sparse autoencoders (SAEs) to break down frozen MRI foundation model embeddings into smaller, more interpretable features. The goal is to understand what different layers encode, and which features are most related to AD progression.
16