Curriculum Vitae

Profile

I am a PhD student specialising in Machine Learning and Medical Image Analysis, with hands-on experience in both academic research and industry. My expertise spans deep learning, computer vision, medical imaging, large-scale ML research, 3D generative AI models, and medical image segmentation.

I am seeking opportunities to apply my expertise in machine learning to real-world challenges in applied research within academia, industry, or startups.

Education

• Ph.D in Machine Learning, University College London, 2021 - Present
  • Supervisors: Prof. John Ashburner, Prof. Cathy J. Price FRS
• MRes in Medical Imaging (Distinction), University College London, 2020 - 2021
  • Key modules: Inverse Problems in Imaging (81%), Machine Learning in Medical Imaging (87%)
• MSci in Chemical Physics (1st Class Honours), University of Bristol, 2016 - 2020

Publications

• L. Chalcroft, I. Pappas, C.J. Price, J. Ashburner, “Synthetic Data for Robust Stroke Segmentation.” Journal of Machine Learning for Biomedical Imaging (MELBA), 2025. [Full text] [Code]

• E. de la Rosa, et al. (including L. Chalcroft), “DeepISLES: a clinically validated ischemic stroke segmentation model from the ISLES’22 challenge.” Nature Communications, 2025. [Full text] [Code]

• L. Chalcroft, J. Crinion, C.J. Price, J. Ashburner, “Domain-Agnostic Stroke Lesion Segmentation Using Physics-Constrained Synthetic Data.” MICCAI, 2025. [Full text] [Code]

• L. Chalcroft, J. Crinion, C.J. Price, J. Ashburner, “Unified 3D MRI Representations via Sequence-Invariant Contrastive Learning.” SASHIMI/MICCAI Workshop, 2025. [Full text]

• L.F. Chalcroft, et al., “LKA: Large-kernel Attention for Efficient and Robust Brain Lesion Segmentation.” 37th Conference on Neural Information Processing Systems (NeurIPS), 2023. [Full text] [Code]

• L.F. Chalcroft, et al., “Development and evaluation of intraoperative ultrasound segmentation with negative image frames and multiple observer labels.” ASMUS/MICCAI, 2021. [Full text] [Code]

Skills

Technical Skills: Python, PyTorch, TensorFlow, JAX, C++, Rust*, MATLAB

Domains: Deep Learning, Computer Vision, Medical Imaging, Segmentation, Spectral Imaging, Remote Sensing, Physics-informed Neural Networks, Neural Operators

*Limited experience

Research Experience

Spectral Computer Vision Scientist (Founding Engineer)

Prospectral, London, UK (Nov 2024 - Present)

• Founding engineer developing novel spectral imaging solutions for agriculture and environmental monitoring
• Leading development of physics-informed neural networks for spectral data analysis
• Implementing neural operators for real-time spectral processing and remote sensing applications
• Building production systems combining computer vision, spectral analysis, and machine learning

Computer Vision Researcher

Tractive, London, UK (Jan 2024 - Oct 2024)

• Led ML research for A16Z-backed pre-seed startup using 3D Generative AI for Retopology
• Translated research from both 3D graphics and generative AI literature
• Conducted large-scale training of transformers using PyTorch and FSDP on Google Cloud VMs
• Wrote production backend code in C++ and Rust

PhD Research Student

University College London, London, UK (Aug 2021 - Present)

• Developed physically-constrained synthetic data framework for robust deep learning in medical imaging
• Created convolutional attention models for 3D medical image segmentation; presented at NeurIPS 2023
• Authored PyTorch library (ssUNet) for 3D contrastive learning in medical imaging
• Applied synthetic data and novel VAE models to stroke lesion segmentation tasks
• Enhanced hypernetworks with self-supervised learning for diverse domain adaptation
• Implemented 3D VD-VAE for normative modeling and anomaly detection in medical imaging

MRes Research Student

University College London, London, UK (Sept 2020 - Aug 2021)

• Developed robust segmentation algorithms using hypernetworks for domain-specific medical imaging
• Created custom PyTorch library for CNNs, hypernetworks, adversarial training, and t-SNE visualisation
• Researched image-level false-positives in segmentation, resulting in publication at MICCAI

Research Scientist (Intern)

Schlumberger Cambridge Research, Cambridge, UK (Aug 2018 - Aug 2019)

• Studied non-newtonian fluids for oil/gas drilling through rheology and diffusing-wave spectroscopy

Leadership & Teaching Experience

• Fellowship Project Supervisor, Fatima Fellowship, London, UK (June 2023 - May 2024)
• MSc Project Supervisor, University College London, London, UK (Jan 2023 - Present)
• Research Supervisor, University College London, London, UK (Sep 2022 - Present)
• Outreach Project Supervisor, In2Research & University College London, London, UK (Aug 2022 - Sep 2022)
• Tutor, Machine Learning and Data Science, Cambridge Spark, London, UK (Jan 2022 - Jan 2024)
• Teaching Assistant, COMP0090: Introduction to Deep Learning, University College London, London, UK (Oct 2021 - Jan 2022)

Grants

• NVIDIA Academic Hardware Grant. Estimated value £5000.
• Google GCP, estimated value $1000.