Automated Pixel-Wise Recalibration for Improving Mass Accuracy and Peak Assignment in MALDI Mass Spectrometry Imaging

Area of Science:

• Analytical Chemistry
• Biotechnology
• Molecular Imaging

Background:

• Mass misalignment is a key limitation in MALDI-TOF MSI, causing poor mass accuracy and spectral broadening.
• This hinders reliable spatial molecular interpretation and data analysis.

Purpose of the Study:

• To develop an automated, pixel-level post-acquisition recalibration framework for MALDI-TOF MSI.
• To improve mass alignment and spectral fidelity without distorting peak shapes.

Main Methods:

• A "one-pixel, one-model" strategy using endogenous calibration features.
• Dynamic peak matching and RANSAC-based regression to correct systematic and heterogeneous mass drift.
• Applied to mouse brain MSI data (33,903 pixels).

Main Results:

• Reduced median absolute mass error from 28.1 to 5.6 ppm.
• Restored spectral compactness and enhanced spatial contrast for lipid visualization.
• Improved resolution of overlapping features and consistency with reference annotation.

Conclusions:

• The pixel-level recalibration framework offers a practical solution for improving mass accuracy and spatial interpretability in MALDI-TOF MSI.
• Demonstrated robustness and transferability across different matrices, ion modes, biological models, and instruments.
• Enables more reliable molecular annotation and analysis on accessible instrumentation.