Neil Oxtoby (Toybox Science)

@alzassociation Trying to work this out… 0.03 SD better off => it takes 2/0.03 = 66.7 years to reach z=2 difference between the two arms (assuming linear). Doesn’t seem impressive. (Simplistic analysis, but please tell everyone how to interpret your effect size & results!)

My team’s latest release of free software for unravelling the mysteries of neurodegenerative diseases is from Isaac Llorente Saguer: joss.theoj.org/papers/10.2110…

Fireside chat at #ADPD2026 including Pharma sees them nodding along with comments about sharing data. I spat out my coffee

@_JakeVogel_ @xiaoyucaly @anlijuncn …start, not stat :-)

@_JakeVogel_ @xiaoyucaly Great stat with @anlijuncn’s excellent talk this morning!

Going to ADPD? Come see our work! We would love to share it with you and chat about it if you are interested. See below for a schedule of our sessions, which cover fMRI, proteomics, transcriptomics and AI

DeepMind at #ADPD2026 claiming that their A.I. Co-Scientist did a particular 10 years of research in 2 days… …did Co-Scientist have to write grants and teach students in parallel?

@anlijuncn @adpdnet @_neuroproteome @biofinder_study Congrats! Great talk today. Hope to see you Thursday, if not before #TheOtherADPD

I’m excited and deeply honored to receive the AD/PD 2026 Junior Faculty Award @adpdnet !!! My sincere thanks to the committee for this recognition. I’m also very grateful to GNPC @_neuroproteome and BioFINDER @biofinder_study for their data support, and especially to my mentor Jacob Vogel @_JakeVogel_ , my brilliant colleagues, and my labmates for their encouragement, generosity, and collaboration. If you’re interested in our work, I’d be delighted to see you at my talk tomorrow morning (Figure credit: GNPC.)

"Do the Evolution", Yield 1998 Watch the full music video at: pj.lnk.to/DoTheEvolution…

@TGO_lab I knew you were famous!

Partilho a minha entrevista para o "HealthNews" onde falo sobre o Prémio Bial em Medicina Clínica, o futuro da investigação em doenças neurodegenerativas e a importância de se investir nas carreiras dos médicos-investigadores. healthnews.pt/2026/02/10/tia…

@PTenigma 9 Hours?!!! Holy Moly. When inspiration hits, eh?

🔥 INCREDIBLE TALK by the LEGENDARY Steve Pizer today !! 🧠 Steve is 84 (!) yet he spoke with extreme mathematical creativity for >9 hours to a packed audience. After earning his BA in Applied Mathematics (1962) and a PhD at Harvard+MIT, he completed what is widely regarded as the first PhD thesis in medical image computing (1967) +founded UNC’s MIDAG lab, training countless innovators in medical image analysis. 🔥 His new work on "evolutionary S-reps" embeds an object-adapted 3D coordinate system inside anatomical objects extracted from brain MRI, to create rich geometric features for disease detection. 🔥 Arbitrarily complex 3D surfaces (eg the cortical sheet) are evolved under conformalized mean curvature flow (CMCF) to an ellipsoid whose interior is then "pulled back" into the original shape, producing a natural intrinsic coordinate system with 3D medial curves + an oriented frame field. 🔥 Mean curvature flow (MCF) was a hot topic in the 1990s for surface parameterization. Kazhdan (JHU) drew on gauge-theoretic ideas - influenced by Karen Uhlenbeck and her team at UTA - to control extreme distortion as the mesh evolved. Yalin Wang (ASU), David Gu, + Shing-Tung Yau published with us (2004) on harmonic mapping of high-genus brain surfaces using holomorphic 1-forms, and Yalin will be visiting Steve and Guorong Wu @UNC this week. 🔥 A clue in the picture hints at how gauge invariance in CMCF can be achieved w/ a conformal basis -> enabling homotopic, bijective mappings between ellipsoids + arbitrarily complex (even high-genus) anatomical surfaces. The whole point of this work is to capture anatomical abnormalities missed by simpler techniques. The proof is left to the reader ;)

A mystery signal is showing up in ADNI amyloid PET scans using NAV-4694. No one knows why. @HAIconference @UCSF ow.ly/tky250Yasbk

Mistaken! There is at least a 4th option, @RussInCheshire — tax wealth, not work. Check out @garyseconomics

@DrNeuroChic I have a similar approach. Got some positive unsolicited feedback after my talk at OHBM: “That was so clear and understandable for non-experts. Thank you.” There are times and places for “technical sophistication” in scientific talks, but a large conference is usually neither…

Unapologetically clear, unapologetically me...how to handle unsolicited feedback I’ve spent years learning how to communicate. I speak the way I do because I listened. I present the way I do because I watched. I teach the way I do because I had to learn differently...and I never assume my audience has it all figured out either. At #AAIC25, I attended an incredible networking event about supporting one another across the unique ways we show up in science...confident or cautious, buttoned-up or boldly balanced. It was meaningful. Uplifting. Which is probably why I was caught off guard when, shortly after, an established professional approached me about a talk I gave the day before...one I had crushed...and offered this unsolicited gem: “Perhaps next time, a bit more sophisticated though.” I froze. Not because I agreed, but because I was reminded how easily #ImposterSyndrome can creep in, even on our best days. Here’s the truth: My presentation style is intentional. I work relentlessly to understand complex science so I can make it accessible to everyone...from my family to frontline clinicians, early-career to established researchers, and curious minds everywhere. Call it simple. Call it casual. Call it whatever you want. But it’s effective, and it’s mine. To those who feel the pressure to sound more “academic,” “technical,” or “sophisticated”... especially when your heart says otherwise: There is power in clarity. There is genius in simplicity. And there is room for your voice in this field. Long hair, big smile, clear words...devoted every day to preventing #Alzheimers and related dementias 💜 #UnapologeticallyUnsophisticated #ScienceCommunication #DrNeuroChic #ScienceWithSoul #ClearOverComplicated

I don't think Twitter works for this anymore, but I had luck with it just <1 year ago, so let's try again! I am recruiting multiple postdoc positions in Alzheimer’s and Dementia, Epidemiology, and Translational Neuroscience! Plz RT! recruit.ap.uci.edu/JPF09486

#AAIC2025 Wednesday POSTER by @DrMaitreiKohli from @POND_UCL & MANIFOLD lab: Wednesday-410. "AutoML-Multiverse: Model Agnostic Framework for Optimizing Individual-level Alzheimer's Disease Classification"

#AAIC2025 Wednesday TALK by @chris_ucl from @POND_UCL: 2pm, session 4-23-DEV. "Longitudinal trajectories of advanced cortical diffusion-weighted imaging measures of tissue microstructure in pre-symptomatic autosomal dominant Alzheimer's disease"

@alzassociation Trying to work this out… 0.03 SD better off => it takes 2/0.03 = 66.7 years to reach z=2 difference between the two arms (assuming linear). Doesn’t seem impressive. (Simplistic analysis, but please tell everyone how to interpret your effect size & results!)

📣 Today is a turning point in what we know about brain health. The results from the U.S. POINTER trial provide the rigorous data needed to say with confidence that healthy behaviors can meaningfully protect brain health across diverse populations in the U.S. The findings, shared for the first time today at #AAIC25, tested lifestyle programs that simultaneously targeted risk factors. The programs focused on: ✅👟Aerobic exercise ✅🧠Cognitive engagement ✅🥗Nutrition ✅🫀Health Monitoring U.S. POINTER is the largest research commitment the Association has made in its history. The landmark results are testament to our commitment to our mission. Thank you to the study participants, our community, our donors, and the staff who made these findings possible. alz.org/USPOINTERresul…

#AAIC2025 Monday POSTER by @LawrenceBinding from @POND_UCL: Monday-437. "Uncovering regional atrophy not explained by tau deposition using Additional Pathology Inference (AddiPath)" A new SuStaIn... :-)

Check out @YoungAlexL’s talk at #AAIC2025 tomorrow. Event-based models with time!

I wish I was with the @POND_UCL team members in Toronto for #AAIC2025. I'm looking forward to hearing all about it