Ninad Aithal

MSc student, Integrated Program in Neuroscience, McGill University.
ORIGAMI Lab, The Neuro (Montreal Neurological Institute), Montréal.

About

I work with the ORIGAMI Lab at the Montreal Neurological Institute (MNI). I work on neuroimaging, medical image analysis and machine learning for brain health. My thesis is on federated learning for Alzheimer's disease progression across imaging sites, where the data stay at each site. I enjoy building open tools that make neuroimaging easier and more fun for everyone.

News

• 29 Oct 2026. NeuroDataHub will be presented at MAIN 2026.
• Oct 2026. NeuroDataHub has had more than 1,200 visitors across 70 countries. Thank you all!
• Oct 2026. Gave a talk at the TOSI open night, "Open Science in Action: Making Neuroimaging Data More Accessible with NeuroDataHub".
• 2026. Attended Brainhack Montréal 2026 and gave an unconference talk, "The butterflies, rainbows and unicorns of vibe coding".
• Jan 2026. Talk and hands-on tutorial on machine learning for neuroimaging, IHUB-Data, IIIT Hyderabad.
• 6–7 Jan 2025. Invited talk and hands-on tutorial (Ripser) on topology for Alzheimer's disease at the National Workshop on Topology for Data Analysis, MITS, Kochi. About 70 attendees.
• Dec 2024. Hands-on tutorial on adaptive deep learning for MRI, SRET winter school.

Projects

I created several open projects, and they are free for everyone to use:

• NeuroDataHub: a catalogue and downloader for public neuroimaging datasets. 50+ datasets, 50,000+ scans.
• NeuroLight: a light PyTorch pipeline for deep learning on MR images. Tutorial.
• NeuroAI Template: a ready-made website for your neuroimaging or brain-AI paper, with interactive 3D brain atlases. Edit Markdown and CSV files, or let an AI agent fill it in from your paper, and publish it free on GitHub Pages. Built with Vite, React, Tailwind, Three.js and NiiVue. Code.
• BrainAge Fingerprints: an ongoing brain age project on explainable brain-age biomarkers.
• Recurrence Plots: an interactive site on detecting mild cognitive impairment from fMRI.
• Getting started with thermal imaging: a site for discovering thermal imaging datasets, with scripts to work with them.

I also contribute to Nipoppy, open and reproducible neuroimaging workflows.

Before

• Research Engineer, Niramai: foundation models for thermal breast imaging.
• Research, Centre for Brain Research, Indian Institute of Science: brain aging and neurodegeneration.
• BS, Indian Institute of Technology Madras. B.Tech, Srinivas University (gold medal).

Papers

• 2026. Data-Driven Identification of Sex Differences in Cerebral Blood Flow Using Arterial Spin Labelling and Explainable Artificial Intelligence · Companion site
• 2026. Data-driven Connectomic Geometry Framework for Multimodal MRI Analysis: Application on Sex Classification
• 2025. Leveraging Persistent Homology for Differential Diagnosis of Mild Cognitive Impairment
• 2025. Persistent homology for MCI classification: a comparative analysis between graph and Vietoris-Rips filtrations
• 2025. Gender and Age-Aware Fully Convolutional Neural Network for Alzheimer's Detection
• 2025. Non-linear dynamics in ECG: a novel approach for robust classification of cardiovascular disorders
• 2024. MCI detection using fMRI time series embeddings of recurrence plots
• 2024. Analyzing Brain Tumor Connectomics Using Graphs and Persistent Homology
• 2024. Multi-scale fMRI time series analysis for understanding neurodegeneration in MCI
• 2024. Simple fully convolutional network to estimate Brain Age
• 2024. Simple fully convolutional network to detect Alzheimer's Disease

All on Google Scholar.

Blog

• Getting Started in NeuroAI Without Spending Months on Preprocessing
• Neuro Data Hub
• Multi-Site MRI data Harmonization
• Intro to AI with Neuroimaging Data: A end to end tutorial using PyTorch
• fMRI preprocessing with FSL
• MRI preprocessing using FSL
• From FMRI to FMRI timeseries
• Recurrence Plots
• Modified 3D Efficient Net
• What makes the analysis of MRI data challenging?

All on Medium.

Contact

Email or LinkedIn is the best way to reach me.

Email · LinkedIn · GitHub · ORCID