IonMorphNet: Generalizable Learning of Ion Image Morphologies for Peak Picking in Mass Spectrometry Imaging

arXiv cs.CV / 4/22/2026

📰 NewsDeveloper Stack & InfrastructureTools & Practical UsageModels & Research

共有:

Key Points

The paper introduces IonMorphNet, a spatial-structure-aware representation model designed to perform peak picking in Mass Spectrometry Imaging (MSI) using fully data-driven learning without task-specific supervision.
To support generalization across acquisition protocols, the authors curate 53 public MSI datasets and train image backbones to classify six structural classes of ion-image spatial patterns.
After training, IonMorphNet can assess ion images and run peak picking without additional hyperparameter tuning, improving performance by +7% mSCF1 over state-of-the-art methods across multiple datasets.
The study also shows that spatially informed channel reduction can enable a 3D CNN for patch-based tumor classification, achieving up to +7.3% Balanced Accuracy compared with pixel-wise spectral classifiers on three tasks.
Code and model weights are released on GitHub, facilitating adoption and further research.

Abstract

Peak picking is a fundamental preprocessing step in Mass Spectrometry Imaging (MSI), where each sample is represented by hundreds to thousands of ion images. Existing approaches require careful dataset-specific hyperparameter tuning, and often fail to generalize across acquisition protocols. We introduce IonMorphNet, a spatial-structure-aware representation model for ion images that enables fully data-driven peak picking without any task-specific supervision. We curate 53 publicly available MSI datasets and define six structural classes capturing representative spatial patterns in ion images to train standard image backbones for structural pattern classification. Once trained, IonMorphNet can assess ion images and perform peak picking without additional hyperparameter tuning. Using a ConvNeXt V2-Tiny backbone, our approach improves peak picking performance by +7 % mSCF1 compared to state-of-the-art methods across multiple datasets. Beyond peak picking, we demonstrate that spatially informed channel reduction enables a 3D CNN for patch-based tumor classification in MSI. This approach matches or exceeds pixel-wise spectral classifiers by up to +7.3 % Balanced Accuracy on three tumor classification tasks, indicating meaningful ion image selection. The source code and model weights are available at https://github.com/CeMOS-IS/IonMorphNet.