Samia Mohinta

I'm a Data Scientist at the University of Cambridge in Albert Cardona's lab, where I develop AI methods and scientific software for mapping connections between neurons in the brains of model organisms.

Over the past few years, I've worked across several areas of machine learning and scientific computing, including medical imaging and clinical data in Parashkev Nachev's High Dimensional Neurology Lab at UCL, and on reinforcement learning with Rui Ponte Costa's Neural & Machine Learning group at the University of Bristol/Oxford. I also had a short stint working on neuromorphic robotics at Bristol Robotics Laboratory. Outside academia, I worked as a Senior Systems Engineer at Infosys, engineering data-driven business intelligence workflows for Apple, and later as Vice President – Innovation at a London-based startup, where I led the development of an NLP-driven recommendation system.

Across these different settings, I've developed a particular interest in building computational workflows that are scalable, reproducible, and useful beyond a single experiment or dataset.

I'm always happy to discuss potential collaborations and consulting opportunities. Feel free to get in touch by email.

Email  /  CV (intermittently updated)  /  Publications  /  Conferences  /  Videos  /  Awards  /  Supervision  /  Service

Portrait of Samia Mohinta

Find me:
GitHub / LinkedIn / Scholar / Twitter

Current Research Interests

AI for Neuroscience & Biomedicine · Reinforcement Learning · Open & Reproducible ML Systems

What's New

  • Oct 2026 Evaluating Contrast-Dependent Errors in Virtual Contrast-Enhanced Brain MRI, with Fatima Fellow Vincess Dongmo and James K Ruffle, accepted at Africa in AI at NeurIPS 2026.
  • Oct 2026 Foreground-Seeded Patch Sampling for 3D Multi-Organelle Segmentation in Volume EM, with Fatima Fellow Muhammad Saleem, accepted as a poster at the NewInML workshop at NeurIPS 2026.
  • Oct 2026 Learning Battery Crack Segmentation from Complete, Partial, and Unlabeled FIB-SEM Data, with Amoghavarsha Mahadevegowda, accepted at ICVGIP 2026.
  • Sep 2026 Predicting brain tumour enhancement from non-contrast MRI with artificial intelligence out in The Lancet Medical Imaging & Theranostics; covered by The Independent.
  • Jul 2026 Paper accepted at the ECCV 2026 BioImage Computing workshop in Malmo, Sweden.
  • Jul 2026 Paper accepted at the IEEE VIS workshop on Visual Analytics in the Age of Autonomous Science.
  • 2026 Selected as a Reproducible Research Fellow at the University of Cambridge.
  • 2026 Talk and poster at Berlin Connectomics Conference.
  • 2026 Paper accepted at OCEANS 2026: Visual Feature Importance in Sim-to-Real Marker Buoy Recognition.
  • 2026 Invited talk at the AIM-Net Biennial Conference, La Rapita, Tarragona.

Publications

For the maintained publication record, please see my Google Scholar.

Brain tumour enhancement prediction from non-contrast MRI thumbnail Predicting brain tumour enhancement from non-contrast MRI with artificial intelligence: a multicohort, retrospective, diagnostic accuracy study
James K. Ruffle, Samia Mohinta, Guilherme Pombo, Asthik Biswas, Alan Campbell, Indran Davagnanam, David Doig, Ahmed Hammam, Harpreet Hyare, Farrah Jabeen, Emma Lim, Dermot Mallon, Stephanie Owen, Sophie Wilkinson, Sebastian Brandner, Parashkev Nachev
The Lancet Medical Imaging & Theranostics, 2026
Media coverage: The Independent / Daily Mail / Oxford Mail / Eastern Daily Press
Fairboard healthcare model equity assessment thumbnail Fairboard: a quantitative framework for equity assessment of healthcare models
James K. Ruffle, Samia Mohinta, Chris Foulon, Mohamad Zeina, Zicheng Wang, Sebastian Brandner, Harpreet Hyare, Parashkev Nachev
arXiv:2604.09656, 2026
Allocentric and egocentric navigation task schematic Hippocampus supports multi-task reinforcement learning under partial observability
Dabal Pedamonti*, Samia Mohinta*, Martin V. Dimitrov, Hugo Malagon-Vina, Stephane Ciocchi, Rui Ponte Costa
Nature Communications, 16, 9619, 2025
*Equal contribution
Synapse proofreading and consensus annotation schematic Beyond Agreement: Standardizing Crowdsourced Synapse Annotations through Proofreading in EM Connectomics
Shi Yan Lee, Ana Correia, Nicolo Ceffa, Miranda Robbins, Daniel Franco-Barranco, Marta Zlatic, Albert Cardona, Samia Mohinta
bioRxiv, 2025
Cross-species synapse detection thumbnail Towards Generalized Synapse Detection Across Invertebrate Species
Samia Mohinta, Daniel Franco-Barranco, Shi Yan Lee, Albert Cardona
arXiv:2509.17041, 2025
Computational limits to brain imaging legibility thumbnail Computational limits to the legibility of the imaged human brain
James K. Ruffle, Robert J. Gray, Samia Mohinta, Guilherme Pombo, Chaitanya Kaul, Harpreet Hyare, Geraint Rees, Parashkev Nachev
NeuroImage, 2024
Brain tumour segmentation with incomplete MRI data thumbnail Brain tumour segmentation with incomplete imaging data
James K. Ruffle, Samia Mohinta, Robert J. Gray, Harpreet Hyare, Parashkev Nachev
Brain Communications, 2023

Conference & Workshop Activity

  • Jun 2025 CATENA presented at Analysis and Modelling of Connectomes, HHMI Janelia, USA.
  • Mar 2025 CATENA presented at AI Revolution Meets 4D Cellular Physiology, HHMI Janelia, USA.
  • 2024 Berlin Connectomics, poster; UK Neural Computation Conference, poster; CATENA presented at Images to Knowledge, Milan, flash talk and poster; UCL NeuroAI Conference, poster.
  • 2023 Selected to participate in the Capocaccia Workshop.
  • 2022 Berlin Connectomics Conference, poster.
  • 2022 S. Mohinta*, D. Pedamonti*, M. Dimitrov, H. Malagon Vina, S. Ciocchi, R. Ponte Costa. Hippocampal networks support continual learning and generalisation. NAISYS, poster.
  • 2022 S. Mohinta*, D. Pedamonti*, M. Dimitrov, H. Malagon Vina, S. Ciocchi, R. Ponte Costa. Hippocampal networks support continual learning and generalisation. COSYNE, poster.
  • 2021 Presented a poster at the Champalimaud Research Symposium.
  • 2021 S. Mohinta*, D. Pedamonti*, M. Dimitrov, H. Malagon Vina, S. Ciocchi, R. Ponte Costa. Hippocampal encoding of continual reinforcement learning features. ACAIN, talk.
  • 2021 S. Mohinta, D. Pedamonti, M. Dimitrov, H. Malagon Vina, S. Ciocchi, R. Ponte Costa. Hippocampal encoding of continual reinforcement learning features. COSYNE, poster.
  • 2020 S. Mohinta, S. Ciocchi, R. Ponte Costa. Hippocampal population representations in reinforced continual learning. COSYNE, poster.
  • 2019 S. Mohinta, S. Ciocchi, R. Ponte Costa. Hippocampal population representations in reinforced continual learning. NeurIPS Biological and Artificial Reinforcement Learning workshop, poster.

Videos & Features

Selected interviews and public features about my work.

Samia Mohinta ThinkLab Hackathon Interview thumbnail
 Samia Mohinta ThinkLab Hackathon Interview
Cambridge ThinkLab interview.
video
Computer Vision in Neuroscience interview thumbnail
 Computer Vision in Neuroscience - Dr Samia Mohinta
Interview with Accelerate Science.
video / LinkedIn post
CATENA feature thumbnail Using AI to map circuitry across entire brains
Public feature about AI for connectomics.
feature

Honors & Awards

  • 2026 Reproducible Research Fellow, University of Cambridge.
  • 2025 Winner (part of a team), CRUK COLDATATHON. Project website.
  • 2024 Dr Teresa Tiffert Research Innovation Award, University of Cambridge.
  • 2023 Oxbridge RisingWISE Women in STEM Fellow.
  • 2023 Second prize, ThinkLab Cambridge / Google / BBC-CAPE hackathon.
  • 2023 Wellcome Data Science Ideathon semi-finalist, mental health challenge for long-term patients. Project GitHub.
  • 2019 Bristol Plus Award.
  • 2018 British Council India 70th Anniversary Scholarship, full tuition waiver for MSc Advanced Computing at the University of Bristol.

Recent Mentoring & Supervision

  • Muhammad Saleem, Fatima Fellow. Our paper Foreground-Seeded Patch Sampling for 3D Multi-Organelle Segmentation in Volume EM was accepted as a poster at the NewInML workshop at NeurIPS 2026.
  • Vincess Dongmo, Fatima Fellow. Our paper Evaluating Contrast-Dependent Errors in Virtual Contrast-Enhanced Brain MRI was accepted at Africa in AI at NeurIPS 2026.
  • Lee Shi Yan [w/ Albert Cardona]. Currently graduate student at Nanyang Technological University (NTU), Singapore.

Service

  • 2026-Present Mentor, Fatima Institute.
  • 2022-Present Mentor, University of Bristol, Year 1 Computer Science and Mathematics students.
  • 2021-2022 Reviewer, Advanced Course and Symposium on Artificial Intelligence and Neuroscience.
  • Oct 2018-Sep 2019 SCEEM Outreach Ambassador for DigiMakers, Bristol University.

Hobbies

Travelling, blogging about moving from India to the UK and student life, and trying new cuisines.