Background
Whole Slide Imaging (WSI) produces gigapixel images, creating major challenges for storage, transmission, and interactive visualization. Current systems rely on classical compression methods such as JPEG, however, these approaches are not fully optimized for WSI-specific characteristics like large homogeneous regions and the need for fast region-of-interest (ROI) access. Balancing compression efficiency with low-latency decoding and smooth navigation remains an open challenge in digital pathology systems.
Objectives
This thesis aims to investigate efficient, compression strategies for WSI.
Key objectives include:
Methodology and Expected Contributions
The work will involve implementing and benchmarking different compression pipelines on WSI datasets. Evaluation will focus on compression ratio, reconstruction quality (PSNR, SSIM), and access performance (latency, ROI decoding). The outcomes are expected to provide practical insights into optimizing WSI compression for both storage efficiency and interactive use.
Expected Student Profile
Type of internship: Master internship
Required educational background: Computer Science, Electrotechnics/Electrical Engineering, Biomedical engineering
Supervising scientist(s): For further information or for application, please contact Saeed Mahmoudpour (< email deleted for security reasons >)
The reference code for this position is 2026-INT-163. Mention this reference code in your application.
Applications should include the following information:
Incomplete applications will not be considered.
Whole Slide Imaging (WSI) produces gigapixel images, creating major challenges for storage, transmission, and interactive visualization. Current systems rely on classical compression methods such as JPEG, however, these approaches are not fully optimized for WSI-specific characteristics like large homogeneous regions and the need for fast region-of-interest (ROI) access. Balancing compression efficiency with low-latency decoding and smooth navigation remains an open challenge in digital pathology systems.
Objectives
This thesis aims to investigate efficient, compression strategies for WSI.
Key objectives include:
- Comparing classical codecs (e.g., JPEG, JPEG2000) for compression efficiency and quality
- Analyzing multi-resolution and tiling strategies for fast ROI access
- Evaluating trade-offs between compression ratio, decoding speed, and usability
Methodology and Expected Contributions
The work will involve implementing and benchmarking different compression pipelines on WSI datasets. Evaluation will focus on compression ratio, reconstruction quality (PSNR, SSIM), and access performance (latency, ROI decoding). The outcomes are expected to provide practical insights into optimizing WSI compression for both storage efficiency and interactive use.
Expected Student Profile
- Programming skills (Python or C/C++)
- Basic image processing knowledge
- Interest in performance evaluation and large-scale data systems
Type of internship: Master internship
Required educational background: Computer Science, Electrotechnics/Electrical Engineering, Biomedical engineering
Supervising scientist(s): For further information or for application, please contact Saeed Mahmoudpour (< email deleted for security reasons >)
The reference code for this position is 2026-INT-163. Mention this reference code in your application.
Applications should include the following information:
- resume
- motivation
- current study
Incomplete applications will not be considered.
