Photo of Patrick Schrempf

Patrick Schrempf

Research Scientist in Multimodal AI for Healthcare

Senior Scientist at Canon Medical Research Europe, working on multimodal learning, clinical NLP, interpretable machine learning, and data-efficient approaches for medical applications.

About me

I am a Senior Scientist at Canon Medical Research Europe and an Honorary Lecturer at the University of St Andrews. My work sits at the intersection of multimodal AI, clinical NLP, and medical imaging, with a focus on practical healthcare applications. I completed an EngD in Computer Science at the University of St Andrews, sponsored by Canon Medical Research Europe, and previously completed a BSc in Computer Science at St Andrews. Originally from Vienna, I grew up bilingual in German and English and now live in Perth & Kinross, working hybrid.

Research

My current research focuses on multimodal learning and clinical NLP for healthcare, with a broader interest in interpretable machine learning and data-efficient learning.

Multimodal learning Clinical NLP Interpretable ML Data-efficient learning

Current roles

Senior Scientist at AI Research, Canon Medical Research Europe.

Honorary Lecturer at the University of St Andrews, affiliated with the AI Research Group and Health Informatics Group.

Highlights

  • MRC Harmonised Impact Acceleration Account funding award of £39,955 through the University of St Andrews in 2025.
  • Best Paper Award at MICCAI LABELS 2020 for "Paying Per-Label Attention for Multi-label Extraction from Radiology Reports".
  • Co-author of a 2026 HeaLing at EACL paper on prompt optimisation for error detection in medical notes.
  • Named inventor on multiple medical AI and language-processing patent filings.

Selected publications

Importance of Prompt Optimisation for Error Detection in Medical Notes Using Language Models

HeaLing Workshop at EACL, 2026

Article | BibTeX

The role of noise in denoising models for anomaly detection in medical images

Medical Image Analysis, Elsevier, 2023

Article | BibTeX

Understanding the performance and reliability of NLP tools predicting stroke phenotypes in radiology reports

Frontiers Digital Health, 2023

Article | BibTeX

Acute stroke CDS: automatic retrieval of thrombolysis contraindications from unstructured clinical letters

Frontiers Digital Health, 2023

Article | BibTeX

Templated Text Synthesis for Expert-Guided Multi-Label Extraction from Radiology Reports

Machine Learning and Knowledge Extraction, 2021

Article | BibTeX

Projects

A few practical tools and experiments alongside my research work.

Panel Timekeeper

A lightweight timer for moderators to keep panel sessions on track across desktop and mobile.

Open Project

COVID-19 Map

An archived interactive map project built around pandemic case-rate data. The original upstream data source has been retired, so this page is kept as a past project rather than a current tool.

View Archive

Curriculum Vitae

A concise overview of my research, academic background, and professional experience.

View CV