Generative deep learning-enabled ultra-large field-of-view lens-free imaging

2024-03-12

arXiv.org

AI Summary

Existing two-dimensional lensless Fourier imaging (LFI) suffers from time-consuming multi-point measurements, complex preprocessing, and high sensitivity to optical parameters, hindering efficient imaging of static thin specimens and precluding real-time, large-field-of-view (FOV) 2D holographic reconstruction for dynamic 3D samples—such as microfluidic droplets or 3D cellular models. To address these limitations, we propose GenLFI, an unsupervised generative AI framework that pioneers the integration of physics-informed generative adversarial networks (GANs) for end-to-end digital hologram reconstruction—eliminating the need for explicit optical modeling or source scanning. GenLFI achieves a record-large FOV of >550 mm²—20× larger than state-of-the-art LFI and 1.76× larger than confocal microscopy—while attaining a resolution of 5.52 μm (sub-pixel level). The method enables model-free, real-time imaging of live cells and 3D microstructures.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Advancements in high-throughput biomedical applications necessitate real-time, large field-of-view (FOV) imaging capabilities. Conventional lens-free imaging (LFI) systems, while addressing the limitations of physical lenses, have been constrained by dynamic, hard-to-model optical fields, resulting in a limited one-shot FOV of approximately 20 $mm^2$. This restriction has been a major bottleneck in applications like live-cell imaging and automation of microfluidic systems for biomedical research. Here, we present a deep-learning(DL)-based imaging framework - GenLFI - leveraging generative artificial intelligence (AI) for holographic image reconstruction. We demonstrate that GenLFI can achieve a real-time FOV over 550 $mm^2$, surpassing the current LFI system by more than 20-fold, and even larger than the world's largest confocal microscope by 1.76 times. The resolution is at the sub-pixel level of 5.52 $mu m$, without the need for a shifting light source. The unsupervised learning-based reconstruction does not require optical field modeling, making imaging dynamic 3D samples (e.g., droplet-based microfluidics and 3D cell models) in complex optical fields possible. This GenLFI framework unlocks the potential of LFI systems, offering a robust tool to tackle new frontiers in high-throughput biomedical applications such as drug discovery.

Problem

• Overcoming time-consuming multi-position measurements in lens-free imaging
• Eliminating strict optical parameterization for large field-of-view imaging
• Enabling real-time 2D imaging for dynamic 3D biomedical samples

Innovation

• Generative unsupervised physics-informed neural network
• Large FOV lens-free imaging setup
• Decoupled reconstruction algorithm from optical parameters