Puti Wen 温菩提

Neuroimaging Scientist | Research Infrastructure | MRI Analysis & Clinical Imaging Tools

Abu Dhabi, UAE pw1246@nyu.edu LinkedIn Interactive Portfolio

A somewhat unconventional neuroscientist. My Ph.D. was in visual neuroscience, but these days I spend most of my time building the infrastructure, pipelines, and practical software that keep brain-imaging projects moving: large MRI datasets, multiple-sclerosis research, and clinician-facing tools. I work comfortably across psychology, radiology, engineering, and clinical neurology, especially when a useful problem doesn’t fit neatly inside one field.

Core Technical Strengths

Infrastructure
XNAT, Slurm/Jubail HPC, Docker, Singularity/Apptainer, GitHub Actions, Box API
Neuroimaging
fMRIPrep, FreeSurfer, TractoFlow, MRIQC, BIDS, dcm2bids, DICOM, ANTs, SPM12, MELD/MAP18
Programming
Python, MATLAB, TypeScript/JavaScript, React Native, Next.js, Node.js/Express, Three.js, Supabase

Experience

Postdoctoral FellowNYU Abu Dhabi, Center for Brain and Health

– Present

Research infrastructure & analysis

  • Imaging data platform. Integrated and deployed leading open-source neuroimaging tools (fMRIPrep, MRIQC, TractoFlow, HCP, dcm2bids, deep-learning visual-area segmentation) as containerized XNAT/HPC pipelines, making them available for researchers to run on their own data directly on the platform; also built two in-house (a BIDS-to-HCP converter and the ARI validator), plus automated data-curation/QA, a nightly validation dashboard, XNAT-to-Box backup, and the documentation the imaging team relies on.
  • Scanner-upgrade QC. Prepared Germany CIMA.X–NYUAD Prisma cross-scanner QC (SNR/tSNR/CNR) and a comparison report that supported the scanner-upgrade decision; also compared NYUAD’s Prisma with UAEU and am conducting pre/post-upgrade QC between the Prisma and newly installed on-site CIMA.X.
  • Normative brain dashboard. Processed 9,000+ HCP and ADNI subjects into an age- and sex-referenced dashboard with an upload-and-compare interface.
  • Stimulus-computer migration. Rebuilt the CBH MRI stimulus setup from Intel to Apple Silicon: ported and tested all active experiments, built reusable experiment templates for future studies, and restored EyeLink eye-tracking and VPixx/ProPixx integration.
  • MS imaging analysis. Co-led tract-based quantitative MRI analysis across 27 white-matter pathways, testing whether lesion location and tissue composition predict MS disability beyond total lesion load.
  • Cross-department fMRI support. Co-author on an in-progress Engineering EEG-fMRI manuscript, provisionally titled “Longitudinal Motor Imagery Training for a Supernumerary Thumb Induces Distinct and Durable Cortical Plasticity,” which examines pre/post-training cortical reorganization during motor-imagery control of a supernumerary thumb.
  • Training & support. Trained 10+ researchers across fMRI, diffusion, vision, biobank, genetics, and MS projects; resolved 25+ reported data-validation and pipeline issues; built afqView, a BIDS-aware browser app for reviewing tractography across 116 subjects.

Clinical imaging tools & support

  • LesionView. Built longitudinal lesion review and clinician-facing HTML/PDF report generation for new and enlarging MS lesions; migrated the original MATLAB prototype to a browser-based app and iterated weekly with Yas Clinic neurologists. Named on the LesionView innovation disclosure (NYU Ref. ABD01-03).
  • CvsView. Central-vein-sign and paramagnetic-rim-lesion review; fixed SWI-FLAIR registration artifacts by relabeling data, built a LightGBM classification workflow, and containerized it for internal research evaluation, using clinical MRI data shared by Harley Street Medical Centre.
  • Brain-atrophy pipeline. Built for internal research review and reporting in collaboration with Yas Clinic clinicians; also ran FLAIR* central-vein-sign processing for Harley Street Medical Centre and Cleveland Clinic Abu Dhabi.
  • Epilepsy MRI. Processed epilepsy MRI at Cleveland Clinic Abu Dhabi using MELD and an in-house MAP18/VBM workflow; built a 1,618-subject reference model, prepared technical reports, and joined an epilepsy multidisciplinary meeting.
  • Presurgical fMRI. Supported task-based fMRI language mapping for an arteriovenous-malformation case: assisted with acquisition, processed and analyzed the data, and reviewed results with the clinical team.

Earlier Research Experience

Ph.D. ResearcherNew York University, Psychology & Center for Neural Science

–
  • Investigated neural pathways for 3D motion perception using stereoscopic stimuli, psychophysics, fMRI, cortical-surface analysis, and multivoxel pattern analysis; developed single-trial decoding methods.
  • Published three first-author papers (Journal of Neuroscience, Imaging Neuroscience, NeuroImage); released code and data through GitHub and OpenNeuro.

Research AssistantVanderbilt University, Randolph Blake Laboratory

–
  • Designed and ran audiovisual-timing experiments; contributed to a first-author publication in Vision (2020).

Education

Ph.D., Cognition and PerceptionNew York University

–

B.S., Psychology; Minor in MathematicsBelmont University

–

Computational Neuroscience: Vision WorkshopCold Spring Harbor Laboratory

Selected Publications & Preprint

Wen, P., Thompson, L. W., Rosenberg, A., Landy, M. S., & Rokers, B. (2025). Single-trial fMRI decoding of 3D motion with stereoscopic and perspective cues. Journal of Neuroscience, 45(22).

Wen, P., Ezzo, R., Thompson, L. W., Rosenberg, A., Landy, M. S., & Rokers, B. (2025). Functional localization of visual motion area FST in humans. Imaging Neuroscience, 3.

Wen, P., Landy, M. S., & Rokers, B. (2023). Identifying cortical areas that underlie the transformation from 2D retinal to 3D head-centric motion signals. NeuroImage, 270, 119909.

Wen, P., Opoku-Baah, C., Park, M., & Blake, R. (2020). Judging relative onsets and offsets of audiovisual events. Vision, 4(1), 17.

Mohamed, A. Z., et al. (incl. Wen, P.) (2026). The ASPIRE Research Institute Dataset. Scientific Data, 13, 186.

Abdullah, O.*, Wen, P.*, et al. (2025). Tract-based quantitative MRI for resolving the clinico-radiological paradox in multiple sclerosis. medRxiv. *Equal contribution.

Intellectual Property & Recognition

  • Co-inventor on one provisional patent application; named on two NYU innovation disclosure applications for LesionView (ABD01-03) and the automated cloud-based MS MRI pipeline / LAMINATE (ABD01-04), submitted through NYU Technology Opportunities & Ventures in May 2026.
  • First Place Poster, NYU Abu Dhabi Division of Science Annual Research Conference, January 2025.
  • Co-author, LAMINATE oral presentation: Best Presentation Award, Tenth MENACTRIMS Congress, December 2025.

Selected Talks & Presentations

  • WHX Dubai Total Radiology Conference: invited talk on AI-powered detection of new lesions and central vein sign in MS; panelist on AI in neuroradiology (Feb 2026).
  • Invited talks and lectures at NYUAD (MS Workshop, CBH Winter Symposium, Neuroimaging Workshop) and the NYU Vision Journal Club (2022–2025); research briefings to NYUAD leadership and a Siemens Healthineers delegation.
  • Posters and talks at Vision Sciences Society (2022–2024), ISMRM (2025), OHBM (2025), MENACTRIMS (2024–2025), and Innovations in Neuroimaging Methods (2026).

Outside the Lab

MyPhysio: physiotherapy practice app · 2026

  • Co-built a full-stack React Native / Next.js / Supabase app with admin dashboards, patient portals, exercise prescription, and pain and range-of-motion tracking.

Shrimp: aquarium simulation game (live) · 2026

  • Browser-based Phaser game with shrimp genetics/DNA testing, multi-tank management, aquascaping, save slots, custom animation, and offline/PWA support.

Languages: English and Chinese; basic Turkish. I design shirts and scientific illustrations for Cold Spring Harbor Laboratory and NYUAD's Center for Brain and Health, make science videos through N.A.R. Things, and keep a self-sustaining aquarium with no filter or routine cleaning. My artwork has appeared in fashion shows in China, and I once reached Grandmaster in Teamfight Tactics.