Related Experiment Video
Updated: Jul 2, 2026
09:31
High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
Published on: April 28, 2022
Hardware-Attentive Programmable Fourier Ptychography Enables Task-Adaptive Label-Free Virtual Staining
Tianyue He1, Wenyi Jing2, Le Zhang3
1College of Electronics and Information Engineering, Sichuan University, Chengdu, Sichuan, China.
Advanced Science (Weinheim, Baden-Wurttemberg, Germany)
|June 30, 2026
Summary
Task-Adaptive Programmable Optics (TAPO) offers reagent-free virtual staining for pathology. This interpretable framework enhances diagnostic accuracy and clinical translation by directly linking optical control to diagnostic goals.
Area of Science:
- Digital pathology
- Computational imaging
- Biomedical optics
Background:
- Virtual staining aims for fast, reagent-free histological imaging for pathology.
- Current methods lack interpretability and cross-task adaptability due to reliance on manual landmarks and blind reconstruction.
- This limits clinical translation of virtual staining technologies.
Purpose of the Study:
- Introduce an interpretable virtual staining framework, Task-Adaptive Programmable Optics (TAPO).
- Enable direct linkage of programmable optical control to task-specific diagnostic objectives, removing manual landmark selection.
- Facilitate task-adaptive multimodal optical encoding for enhanced virtual staining.
Main Methods:
- Utilize a specially modulated visible light imaging system.
- Employ an optics-attention-based reconstruction approach.
- Implement a programmable imaging system that decomposes optical steps for task-adaptive encoding.
Main Results:
- TAPO demonstrated successful validation across diverse tissues (liver, lung) and pathological states (normal, malignant).
- The framework proved effective with different staining protocols (H&E, Masson's trichrome) and magnifications (10×, 20×).
- Pathologist assessments showed high clinical acceptance regarding resolution, cytoplasmic clarity, and nuclear staining, even with varied section thicknesses (4 µm, 10 µm).
Conclusions:
- TAPO provides an interpretable and adaptable virtual staining solution.
- The framework shows significant potential for clinical pathology practice.
- TAPO overcomes limitations of previous virtual staining methods, paving the way for wider adoption.
