Sammy Shorthouse

‌

Email: sammy.shorthouse@bristol.ac.uk 

LinkedIn: https://www.linkedin.com/in/sammy-shorthouse/

GitHub: https://github.com/SammySho 

Project title: A machine learning platform for the personalised treatment of ovarian cancer

Supervisory team: James Armstrong (primary), Qiang Liu (secondary), Aya Elghajiji and Nathan Lepora (other)

 

Project summary

My PhD research focuses on developing computational and imaging approaches to better evaluate three-dimensional cell culture models, such as spheroids and organoids. By combining advanced image analysis with machine learning, the project aims to create scalable, non-destructive methods for assessing tissue morphology and treatment response. This work supports more reliable use of organoid systems in biomedical research and drug discovery.

Bio

In June 2020, I graduated from the University of Sussex with a BSc in Computer Science. Towards the end of my studies, I realised I wanted to learn more about artificial intelligence and applicable research within the field. My final year project at Sussex focused on multi-agent reinforcement learning and designing environments to encourage cooperation and competition. I enjoyed this project but felt I wanted to do research that had a more immediate impact.

I then studied an MSc in Artificial Intelligence at King's College London to deepen my understanding of machine learning techniques. My dissertation used Gaussian processes to predict the diagnosis of sepsis from intensive care unit data, which developed into an exciting and challenging project.

Since then, my work has focused on developing machine learning methods for complex and noisy datasets, with applications across healthcare, biology, and beyond. I am particularly interested in how data-centric AI can be used to generate meaningful insights, support decision-making, and bridge the gap between theoretical methods and real-world impact.

Research and Activity

  • 2025 - TCES 2025 and ICSS 2025: Deep learning + data labelling for neural organoids - Speaker
  • 2025 - ISMB 2025 and AIBIO UK 2025: Prediction of organoid morphology using deep learning - Poster presentation
  • SegmentWise- SegmentWise is a professional annotation tool designed to streamline segmentation of biological images such as organoids and cells. The application integrates manual drawing, threshold-based segmentation, and AI-assisted workflows using the Segment Anything Model (SAM). It supports experiment-based image organisation, annotation progress tracking, and efficient batch operations, enabling researchers to generate high-quality segmentation masks with both precision and scalability.