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

2026-04-21

AI Summary

This work addresses the challenge of peak extraction in mass spectrometry imaging (MSI), which typically relies on dataset-specific hyperparameters and struggles to generalize across acquisition protocols. The authors propose IonMorphNet, the first unsupervised peak extraction framework that explicitly models the spatial structure of ion images. Leveraging 53 publicly available MSI datasets, they construct six spatial morphology labels to train image backbone networks—such as ConvNeXt V2-Tiny—for structural classification. This spatial-aware representation is then integrated with channel dimensionality reduction to guide peak selection without task-specific supervision. The method achieves fully automated, cross-dataset generalizable peak extraction, improving mean spatially constrained F1 (mSCF1) scores by up to 7% across multiple datasets. When applied to tumor classification tasks, it attains balanced accuracy gains of up to 7.3%, matching or surpassing pixel-level spectral classifiers.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Natural Language Processing: Information Extraction

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Bridging structured and unstructured data
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

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.

Problem

• peak picking
• Mass Spectrometry Imaging
• ion image
• generalization
• hyperparameter tuning

Innovation

• IonMorphNet
• mass spectrometry imaging
• peak picking
• spatial morphology
• generalizable representation