Liam Chalcroft

Software

Everything below is on GitHub under a permissive licence.

Libraries

  • medrs

    Rust · Python

    Medical imaging I/O and processing for throughput-critical deep learning pipelines. Memory-mapped NIfTI reading, crop-first loading of sub-volumes, and a lazily evaluated transform pipeline that fuses consecutive axis operations. The repository is named med-rs and the package is published as medrs on crates.io and PyPI.

  • gaze

    Python

    Agentic framework for medical vision-language models: viewer-level image tools, multi-turn tool use, and literature retrieval from PubMed and Open-i. Published on PyPI as gaze-vlm.

Models and paper code

  • SynthStroke

    Python · MELBA 2025

    Code and pretrained weights for stroke lesion segmentation trained entirely on synthetic data, with no sequence-specific training images. Also packaged as an SPM12 toolbox for use without a Python environment.

  • qsynth

    Python · MICCAI 2025

    Physics-constrained synthetic qMRI generation for domain-agnostic lesion segmentation. Implements qATLAS, which estimates qMRI maps from MPRAGE, and qSynth, which synthesises them directly from tissue labels.

  • contrast-squared

    Python · SASHIMI 2025

    Sequence-invariant contrastive pre-training for 3D MRI. A single encoder transfers across segmentation and denoising tasks and across sites. Weights are on Hugging Face.

  • MDUNet

    Python · NeurIPS 2023 workshop

    All-convolutional transformer block for 3D brain lesion segmentation, giving transformer-like long-range modelling at CNN parameter cost.

  • RectAngle

    Python · MICCAI 2021 workshop

    Segmentation and classification for trans-rectal B-mode ultrasound, including the pre-screening classifier that suppresses false positives on frames with no region of interest.

Upstream contributions