pyM2aia

pyM2aia provides memory-efficient access to and processing of mass spectrometry imaging (MSI) data in imzML format to enable computational analysis of large 3D and multi-modal MSI datasets.

Key Features:

• Memory-Efficient Data Handling: Optimized for processing large MSI datasets stored in imzML format to minimize system memory usage.
• Batch Generator Utilities: Includes batch generator utilities to support batch processing and training workflows for deep learning on large MSI datasets.
• Signal Processing Workflows: Supports signal processing workflows for MSI data, including image segmentation and related processing tasks.
• Deformable 3D Image Reconstruction: Provides deformable 3D image reconstruction methods for volumetric MSI datasets.
• Multi-Modal Registration: Implements multi-modal registration techniques to align datasets across imaging modalities.
• Multi-Modal Data Integration: Enables fused-data representation with individual mass axes within a shared coordinate system for integrated analysis.
• MITK Integration: Implemented as an extension of MITK (Medical Imaging Interaction Toolkit) for medical image processing interoperability.

Scientific Applications:

• N-glycan Mouse Kidney Reanalysis: Applied to reanalysis of an N-glycan mouse kidney MSI dataset.
• Mouse Brain Lipid and Peptide Studies: Applied to 3D reconstruction and multi-modal image registration of lipid and peptide datasets from a mouse brain.
• Spatial Biomolecule and Pharmaceutical Mapping: Used to analyze spatial distributions of biomolecules and pharmaceuticals within tissue specimens.

Methodology:

Computational methods explicitly include memory-efficient access and processing of imzML MSI data, batch generator utilities for deep learning, signal processing workflows including image segmentation, deformable 3D image reconstruction, multi-modal registration, and fused-data representation with individual mass axes; implemented as an extension of MITK.