DPC-QA Net: A No-Reference Dual-Stream Perceptual and Cellular Quality Assessment Network for Histopathology Images

2025-09-19

AI Summary

Whole-slide imaging (WSI) is frequently degraded by reference-free artifacts—including staining inconsistencies, defocus, and cellular degradation—posing significant challenges for automated analysis. To address this, we propose a dual-stream no-reference image quality assessment (NR-IQA) network that jointly models global perceptual quality and cell-level structural integrity: a global branch leverages wavelet-domain discrepancy modeling, while a cellular branch incorporates nuclear membrane embedding representations. Cross-attention fuses features from both streams, and an Aggr-RWKV module enhances local–global feature correlation. The model is trained end-to-end with a multi-task loss. Evaluated on multiple WSI datasets, it achieves >92% accuracy in detecting staining and cellular abnormalities. It also significantly outperforms state-of-the-art NR-IQA methods on general-purpose benchmarks (LIVEC and KonIQ). Crucially, its predicted quality scores exhibit strong correlation with downstream cell recognition performance, demonstrating clinical utility.

Technology Category

• Computer Vision: Medical and Biological Imaging
• Knowledge Representation and Reasoning: Qualitative Reasoning
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content

Abstract

Reliable whole slide imaging (WSI) hinges on image quality,yet staining artefacts, defocus, and cellular degradations are common. We present DPC-QA Net, a no-reference dual-stream network that couples wavelet-based global difference perception with cellular quality assessment from nuclear and membrane embeddings via an Aggr-RWKV module. Cross-attention fusion and multi-term losses align perceptual and cellular cues. Across different datasets, our model detects staining, membrane, and nuclear issues with >92% accuracy and aligns well with usability scores; on LIVEC and KonIQ it outperforms state-of-the-art NR-IQA. A downstream study further shows strong positive correlations between predicted quality and cell recognition accuracy (e.g., nuclei PQ/Dice, membrane boundary F-score), enabling practical pre-screening of WSI regions for computational pathology.

Problem

• Detects staining artifacts and cellular degradations in histopathology images
• Assesses global perceptual quality and cellular-level features simultaneously
• Evaluates WSI usability for computational pathology through quality-correlated metrics

Innovation

• Wavelet-based global difference perception
• Cellular quality assessment via Aggr-RWKV
• Cross-attention fusion with multi-term losses