Abstract
Co-registering whole-slide histology images (WSIs) with mass
spectrometry imaging (MSI) can enable morphology-aware molecular
analysis and annotation transfer for computational pathology. Landmarkfree
registration is faced with a fundamental challenge: MSI is a sparse,
high-dimensional spectral image, so conventional image-registration pipelines
require an informative MSI surrogate.We present DreaMS-Align, a landmarkfree
WSI-MSI co-registration workflow that generates this surrogate from
zero-shot embeddings of a pretrained mass spectrometry foundation model,
avoiding manual ion selection, labels, and retraining. The embeddings
can be projected and thresholded to generate a robust mask for tissue
boundaries, suitable for contour-based registration methods. We implement
such a workflow in imzDesk, an open-source self-hosted web-based
workbench and Python library. We evaluate this workbench on 109 public
WSI-MSI pairs, and DreaMS-Align achieves the lowest median ASSD
(180 μm), with a median runtime of 46 s per pair.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/COMPAYL_067.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
BibTex
@InProceedings{FarMoh_DreaMSAlign_MICCAISAT2026,
author = { Farahmand, Mohammad AND Jamzad, Amoon AND Policelli, Robert AND Mousavi, Parvin},
title = { { DreaMS-Align: Zero-Shot Spectrum-Embedding Surrogates for Scalable WSI-MSI Co-registration } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17251},
month = {pending},
page = {pending}
}