Computational Reconstruction for Lensless Structured Illumination Microscopy

Imec; Vlaams-Brabant; Sur place; EUR 8 900 - 16 000; Plein temps

Résumé du poste

Imec is offering a Master internship in Leuven focused on computational reconstruction for lensless structured illumination fluorescence imaging. The project aims to design and evaluate advanced reconstruction algorithms that convert multiplexed measurements into high-resolution images for high-throughput assays.

The internship explores the interplay between forward modelling and image reconstruction, with tasks including algorithm development, dataset generation, and defining image quality

Qualifications

• Master's student in CS, EE, or Physics.
• Background in image reconstruction or inverse problems is desirable.
• Experience with datasets and reconstruction concepts is helpful.

Responsabilités

• Investigate and optimize the forward model and reconstruction metrics.
• Develop the computational pipeline turning measurements into quantitative images.
• Perform forward modelling, algorithm and network development, and dataset generation.
• Define application-oriented image quality metrics for high-throughput biology.

Connaissances

Forward modelling

Algorithm development

Network development

Dataset generation

Image quality metrics

Formation

Master's in Computer Science

Master's in Electrical Engineering

Master's in Physics

Description du poste

Computational Reconstruction for Lensless Structured Illumination Microscopy

Computational Reconstruction for Lensless Structured Illumination Microscopy

Master internship - Leuven | More than two weeks ago

Design and evaluate advanced reconstruction algorithms that turn compact lensless structured-illumination measurements into high-throughput, high-resolution fluorescence images for next-generation biological assays.

Lensless structured illumination microscopy offers a promising route to overcome the space-bandwidth limitations of conventional objective-based systems, enabling large fields of view and enhanced resolution in a compact architecture. By combining tailored illumination patterns with lens-free detection and computational reconstruction, such systems are well suited for high-throughput fluorescence imaging across diverse assay formats, including microfluidic chips and multi-well plates. At the same time, the resulting inverse problem couples illumination, propagation, sensor sampling, and noise, calling for advanced reconstruction strategies that balance image quality, robustness, and computational efficiency.

This internship will focus on investigating and optimizing the interplay between the forward model and image reconstruction metrics for lensless structured illumination fluorescence imaging. The work will focus on the computational pipeline that transforms multiplexed raw measurements into quantitatively reliable images, which is rooted in the fundamental physics of image formation, inverse problem solving and information theory . Depending on the student’s background, tasks may include forward modelling, algorithm and network development, dataset generation, and the definition of application-oriented image quality metrics for high-throughput biological measurements.

Type of internship: Master internship

Required educational background: Computer Science, Electrotechnics/Electrical Engineering, Physics

The reference code for this position is 2026-INT-054.

Lensless structured illumination microscopy offers a promising route to overcome the space-bandwidth limitations of conventional objective-based systems, enabling large fields of view and enhanced resolution in a compact architecture. By combining tailored illumination patterns with lens-free detection and computational reconstruction, such systems are well suited for high-throughput fluorescence imaging across diverse assay formats, including microfluidic chips and multi-well plates. At the same time, the resulting inverse problem couples illumination, propagation, sensor sampling, and noise, calling for advanced reconstruction strategies that balance image quality, robustness, and computational efficiency.

This internship will focus on investigating and optimizing the interplay between the forward model and image reconstruction metrics for lensless structured illumination fluorescence imaging. The work will focus on the computational pipeline that transforms multiplexed raw measurements into quantitatively reliable images, which is rooted in the fundamental physics of image formation, inverse problem solving and information theory . Depending on the student’s background, tasks may include forward modelling, algorithm and network development, dataset generation, and the definition of application-oriented image quality metrics for high-throughput biological measurements.

Type of internship: Master internship

Required educational background: Computer Science, Electrotechnics/Electrical Engineering, Physics

Supervising scientist(s): For further information or for application, please contact Quentin Desmeth ( [e-mail] ) and Steven Vanuytsel ( [e-mail] )

The reference code for this position is 2026-INT-054.

imec's cleanroom