MxIF Q-score: Biology-Informed Quality Assurance for Multiplexed Immunofluorescence Imaging
Shunxing Bao1, Jia Li2, Can Cui3
1Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, USA.
Medical Optical Imaging and Virtual Microscopy Image Analysis : First International Workshop, MOVI 2022, Held in Conjunction with MICCAI 2022, Singapore, September 18, 2022, Proceedings. MOVI (Workshop) (1St : 2022 : Singapore)
|December 29, 2025
Summary
We developed a novel learning-based quality score for multiplex immunofluorescence (MxIF) imaging. This automated system ensures high-quality spatial single-cell data analysis, crucial for accurate biological insights.
Area of Science:
- Computational pathology
- Bioinformatics
- Digital pathology
Background:
- Whole slide imaging (WSI) presents significant challenges due to its high resolution.
- Multiplex immunofluorescence (MxIF) further complicates analysis by combining dozens of WSIs per tissue.
- Manual data curation for MxIF is infeasible, risking misleading biological findings.
Purpose of the Study:
- To introduce an automated, biology-informed quality scoring system for MxIF image data.
- To address the need for reliable quality assurance in complex spatial single-cell analyses.
- To improve the rigor of downstream analyses like cell type annotation.
Main Methods:
- Developed a learning-based MxIF quality score (MxIF Q-score).
- Integrated automatic image segmentation and single-cell clustering.
- Employed a biology-informed approach for data curation.
Main Results:
- The MxIF Q-score system achieved 0.99 recall and 0.86 precision for spatial alignment validation against manual checks.
- Validated on 245 MxIF image regions of interest (ROIs) across 49 WSIs.
- Demonstrated efficacy through extensive experimental results.
Conclusions:
- The proposed MxIF Q-score provides automatic quality assurance for image alignment and segmentation.
- This is the first study offering an automated, biologically informed quality score for MxIF data.
- A biology-knowledge-driven scoring framework is a promising approach for assessing complex MxIF data.
