psudo: Exploring Multi-Channel Biomedical Image Data with Spatially and Perceptually Optimized Pseudocoloring

Area of Science:

• Biomedical imaging
• Computational biology
• Data visualization

Background:

• Multichannel fluorescence imaging is crucial for visualizing protein spatial relationships in biological samples.
• Digital visualization software allows flexible pseudocoloring and channel combination for data exploration.
• Optimizing color palettes is essential for accurate interpretation of complex multichannel data.

Purpose of the Study:

• To introduce psudo, an interactive system for generating optimal color palettes for multichannel spatial data.
• To develop a novel optimization method that maximizes perceptual differences between image channels and minimizes color blending.
• To provide users with tools for exploring multichannel image data and evaluating color palettes.

Main Methods:

• A novel optimization algorithm generates color palettes maximizing inter-channel perceptual differences.
• An interactive system integrates this method for user exploration and palette evaluation.
• An interactive lensing approach offers on-demand feedback on channel overlap and color confusion metrics.

Main Results:

• The psudo system allows users to create and apply color palettes globally or locally.
• A crowdsourced user study with 150 participants demonstrated improved accuracy in data discernment and comparison using the proposed approach.
• The system was successfully applied to analyze complex immune responses in cancer tissue data.

Conclusions:

• The psudo system effectively optimizes color palettes for multichannel fluorescence imaging, enhancing data interpretability.
• The developed optimization method and interactive tools improve users' ability to analyze spatial protein relationships.
• This approach has significant potential for advancing biological research, particularly in fields like cancer immunology.