Abstract
The purpose of this research is to develop efficient optical microscopy techniques for large-scale SR imaging with high speed and high photon efficiency. By implementing a continuous scanning strategy, high-speed image acquisition on a large-scale is realized. Furthermore, real-time digital pixel reassignment and instant optical photon reassignment are performed to enhance the resolution by a factor of ~1.4×. Additionally, the flexibility of the scanning strategy allows for easy implementation of background subtraction based on the hybrid illumination and line modulation principle, ensuring superior resolving ability even in dense or thick biological samples.Image scanning microscopy (ISM) could nearly double resolution and partially reject out-of-focus background signals without in-focus photon loss. The key principle of ISM is to reassign the signals that are recorded by array detectors, thereby reconstructing the effective point spread function of the system. Traditional point ISM, utilizing single-point excitation to scan over the sample via galvos, suffers from limited imaging speed. Parallelized ISM, utilizing structured multi-point or multi-line illumination generated by optics such as a digital micromirror device and spinning disk, enables higher acquisition speed at the expense of system and computational complexity. Moreover, while the imaging scale is larger than a single FoV, the stop-stare imaging stitching process harnesses imaging efficiency. Additionally, though ISM shows superiority in imaging thick samples with SR due to its robustness and high photon efficiency, the increased background from the defocus plane still degrades image quality.
To address this issue, we propose a versatile pixel-reassigned continuous scanning microscopy, including the digital pixel-reassigned line-scanning microscopy (dPRLM) for real-time SR reconstruction, and the continuous image scanning microscopy (cISM) with optical reassignment for instant SR imaging on a large scale. Specifically, in dPRLM, two fixed orthogonal line illuminations and line-array detection (24 kHz for 8 × 2048 pixels, 12 kHz for 16 × 2048 pixels) are employed to continuously scan and image the samples in two orthogonal directions, respectively. Then, computational reassignment of the pixels in recorded line images to the fixed line-excitation center at each scanning position is performed for resolution enhancement. Next, the SR information from the two scanning directions is fused using wavelet-based fusion to obtain isotropic lateral resolution. Finally, a modified Hybrid illumination (HiLo) algorithm is developed to remove the out-of-focus background signal, further preserving the in-focus SR information. Instead of computational processing, in cISM, a fast polygon scanner (i.e., 480 Hz × 36 facets = 17.28 kHz) is utilized to generate virtual line-shaped excitation, meanwhile acting as a de-scanner and re-scanner to reassign the emission photon in the line direction. Optical pixel reassignment could be instantaneously achieved along this scan axis by inserting a beam expander within the de-scanning and re-scanning. Meanwhile, orthogonal to the line direction, the sample moves continuously by the stage in synchronization with the camera works in sub-array mode (16 × 2400 pixels), enabling automatically pixel reassignment of each frame whiling performing summation in the time delay integration (TDI) approach.
Scalar parametric models have been developed to simulate imaging results. Imaging experiments were designed and performed to verify the predicted performance on various biological samples, which show an over 1.4× resolution enhancement in the lateral direction and over 1.5× enhancement in the x direction. Fast imaging speed at video rate (line rate over 12 kHz) can be achieved without a field limit along the continuous scanning direction. These results verify that pixel-reassigned continuous scanning microscopy is a facile and powerful tool for visualizing large-area samples with SR, which may find critical applications in biophotonics and industry inspections.
| Date of Award | 25 Sept 2025 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Wai Yan Kannie CHAN (Supervisor) & Shih-Chi Chen (External Co-Supervisor) |