Master’s Thesis: Finding the Best Areas for Automated Blood Cell Analysis

Reliable automated blood cell analysis depends on selecting appropriate regions of a blood smear. This thesis at CellaVision investigates how characteristics used by morphologists can inform an explainable, automated assessment of region suitability. The project starts with high-resolution images and explores how similar assessments can be achieved at lower resolution for efficient scanning.

The assignment

• Review relevant research and/or consult morphologists and other specialists to establish how suitable blood smear regions can be assessed.
• Identify and evaluate useful image features, including red blood cell overlap, central pallor, cell distribution and empty background.
• Develop and test an explainable image-analysis approach with machine learning as a core component.
• Investigate how findings from high-resolution images can support efficient assessment of low-resolution images.
• Compare results with existing methods and, where feasible, expert morphologists’ assessments.
• Aim to deliver a prototype and findings that could support improvements in automated blood cell analysis.

Who we are looking for

• One or two students studying for a master’s degree in Computer Science, Engineering Mathematics, Biomedical Engineering or a related discipline.
• Experience with machine learning and an interest in programming and analytical problem-solving.
• An interest in image processing and computer vision is an advantage.

What we offer

• Practical experience addressing a medical imaging problem alongside experienced engineers.
• Close collaboration with specialists and leaders, providing insight into advanced technology and the operations of an international company.
• Opportunities to build professional connections and explore a potential future career at CellaVision.
• A modern headquarters office with good infrastructure and commuting connections.

Practical details

Master’s thesis worth 30 ECTS, planned for spring 2027. Based at CellaVision’s headquarters in Lund, for one or two students. Apply through the application form with a cover letter and transcript of records by November 30, 2026.