Optimizing Paths for Adaptive Fly-Scan Microscopy: An Extended Version
Thomas F. Lynn, Ming Du, Zichao Di, Sven Leyffer · 2025
Read Optimizing Paths for Adaptive Fly-Scan Microscopy: An Extended Version by Thomas F. Lynn, Ming Du, Zichao Di, Sven Leyffer (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-01869.txt.
Introduction
Yu Lu leads four co-authors on smarter scan paths for fly-scan X-ray microscopy. In fly-scan mode, the probe moves without stopping, cutting scan time versus step by step grids. Not all image regions matter equally, yet standard fly-scans treat them alike. The team instead steers the scan toward regions of interest and fills the rest by image completion.
Each round scores the current image to place anchor points near promising regions. An objective then tunes those points, and the shortest path links them for the next scan. Fresh data update the picture before the next round begins. The loop shortens scans and may lower X-ray dose while keeping key detail sharp. It also drops the need for large training sets.
Catalog record
| CREATOR | Thomas F. Lynn, Ming Du, Zichao Di, Sven Leyffer |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-01869
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Optimizing Paths for Adaptive Fly-Scan Microscopy: An Extended Version Thomas F. Lynn, Ming Du, Zichao Di, Sven Leyffer Abstract In x-ray microscopy, traditional raster-scanning techniques are used to acquire a microscopic image in a series of step-scans. Alternatively, scanning the x-ray probe along a continuous path, called a fly-scan, reduces scan time and increases scan efficiency. However, in many imaging scenarios, not all regions of an image are equally important. Currently used fly-scan methods do not adapt to the characteristics of the sample during the scan, often wasting time in uniform, uninteresting regions. While we are not aware of any adaptive solution for fly-scan methods, one approach to avoid unnecessary scanning in uniform regions for raster step-scans is to use deep learning techniques to select a shorter optimal scan path instead of a traditional raster scan path, followed by reconstructing the entire image from the partially scanned data. However, this approach heavily depends on the quality of the initial sampling, requires a large dataset for training, and incurs high computational costs. We propose leveraging the fly-scan method along an optimal scanning path, focusing on regions of interest (ROIs) and using image completion techniques to reconstruct details in non-scanned areas. This approach further shortens the scanning process and potentially decreases x-ray exposure dose while maintaining high-quality and detailed information in critical regions. To achieve this, we introduce a multi-iteration fly-scan framework that adapts to the scanned image. Specifically, in each iteration, we define two key functions: (1) a score function to generate initial anchor points and identify potential ROIs, and (2) an objective function to optimize the anchor points for convergence to an optimal set. Using these anchor points, we compute the shortest scanning path between optimized anchor points, perform the fly-scan, and subsequently apply image completion based on the acquired information in preparation for the next scan iteration. GOVERNMENT LICENSE The submitted manuscript has been created by UChicago Argonne, LLC, Operator of Argonne National Laboratory (“Argonne”). Argonne, a U.S. Department of Energy Office of Science laboratory, is operated under Contract No. DE-AC02-06CH11357. The U.S. Government retains for itself, and others acting on its behalf, a paid-up nonexclusive, irrevocable worldwide license in said article to reproduce, prepare derivative works, distribute copies to the public, and perform publicly and display publicly, by or on behalf of the Government. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan. http://energy.gov/downloads/doe-public-access-plan. I Introduction Scanning microscopy is an imaging technique where a focused beam (of x-rays or electrons, for example) is scanned across the surface of the sample being measured. At each scanned location, a detector collects the measured intensity of the emitted signal from the sample. Conventionally, this technique is conducted in as a series of ”step-scans”: for each scanned point, the sample stage needs to come to a full stop before signals are collected while the sample is stationary. The stage then accelerates from static and moves to the next point. Compared to the traditional step-scan, a new method called ”fly-scan” has been introduced and has been popular for its high-speed scanning capabilities [ 1 , 2 , 3 ] . In fly-scan, the sample stage moves constantly, and signals are collected as the sample is in motion. Although this introduces motion blur, ultra-bright light sources, such as synchrotron facilities that generate high-flux x-ray beams, can yield sufficient signals within a short exposure window, thereby keeping the motion blur well under control. In fly-scan, four critical concepts are central to its operation: exposure time, dead time, scan speed, and beam size. The exposure time is the period during which the detector actively collects data; the detector reports the integrated signal intensity during each exposure window as a readout value. The dead time is the time when no data is collected due to processing or hardware limitations. With state-of-the-art techniques, the dead time is typically much shorter compared to the exposure time [ 4 ] , enhancing the efficiency of data acquisition. The scan speed is the velocity at which the scanning system moves across the sample. The speed directly affects spatial resolution and integration time. Faster motion reduces the sampling time at each point, but can potentially compromise scan quality. Lastly, the beam size is the dimensions of the probe or beam (e.g., X-ray or laser) used to interact with the sample. Fig. 1 illustrates these key components of fly-scan. Fig. 1: A schematic diagram illustrating the key components of fly-scan microscopy. Assuming the probe has a circular footprint on the sample (gray dashed circle), it sweeps the area enclosed by the green line during each exposure time t_{e} . The signals integrated during this window constitute a detector readout. Between exposure windows, there is a short dead time t_{d} where the detector processes and sends out the measured value, during which no signal is collected. In many imaging scenarios, not all regions of an image are equally important. Using a raster fly-scan path, where the probe or sample stage moves in a regular pattern so that the probe visits the entire sample surface, often results in unnecessary scanning of uniform regions. To address this, this paper proposes an adaptive fly-scan framework that prioritizes scanning regions of interest (ROIs) along a specific designed optimal scan path, while employing image completion techniques to reconstruct details in non-scanned areas. Using this proposed technique, the scan time can be reduced to one-third of the original time, while maintaining high-quality final images after image completion. Specifically, the main contributions of our work are summarized as follows: • We propose an adaptive fly-scan framework that utilizes less than 30\% of the measurements in dense scan while maintaining high-quality final images through an image completion technique. • We define an efficient score function and objective function to generate optimal anchor points for scan routing. • We design three experiments to verify the effectiveness of the proposed functions and demonstrate the generalization capability of the proposed framework. II Related Works Beyond the basic concepts of fly-scan, it is important to review adaptive scanning and image completion techniques. II-A Adaptive Scanning In general, scanning approaches can be categorized into raster scanning [ 5 ] and adaptive scanning [ 6 ] , depending on whether the full image or only a portion of it is scanned. Additionally, scans can be classified as fly-scan [ 7 ] or step-scan [ 8 ] based on whether the scanning motion is continuous or discrete. Raster scanning is the baseline approach widely used in research. Some variant methods of raster scanning use space-filling curves, instead of a rectangular scan grid, to maximize the coverage of the probe in the sample surface, such as specially designed raster scan paths [ 9 ] or varying scan speeds across different areas of the image [ 10 ] . Adaptive scanning was later introduced for use with step-scan techniques [ 6 ] . More recently, deep learning methods have been applied to adaptive step-scan frameworks [ 11 ] , achieving impressive results [ 12 ] . However, to the best of our knowledge, these adaptive scanning approaches have not yet been extended to fly-scan. This serves as