DM-QPMNET: Dual-modality fusion network for cell segmentation in quantitative phase microscopy

TL;DR

DM-QPMNet is a dual-encoder neural network that effectively fuses polarized intensity images and phase maps for improved cell segmentation in quantitative phase microscopy, outperforming traditional and simple deep learning methods.

Contribution

The paper introduces a novel dual-encoder architecture with multi-head attention for modality-specific feature fusion in ssQPM cell segmentation.

Findings

Significant improvement over baseline methods.

Effective multi-modal feature integration.

Robust segmentation performance demonstrated.

Abstract

Cell segmentation in single-shot quantitative phase microscopy (ssQPM) faces challenges from traditional thresholding methods that are sensitive to noise and cell density, while deep learning approaches using simple channel concatenation fail to exploit the complementary nature of polarized intensity images and phase maps. We introduce DM-QPMNet, a dual-encoder network that treats these as distinct modalities with separate encoding streams. Our architecture fuses modality-specific features at intermediate depth via multi-head attention, enabling polarized edge and texture representations to selectively integrate complementary phase information. This content-aware fusion preserves training stability while adding principled multi-modal integration through dual-source skip connections and per-modality normalization at minimal overhead. Our approach demonstrates substantial improvements over monolithic concatenation and single-modality baselines, showing that modality-specific encoding with learnable fusion effectively exploits ssQPM's simultaneous capture of complementary illumination and phase cues for robust cell segmentation.