Georgia Institute of Technology
Large-Field High-Temporal-Resolution Microscopic Imaging: Deconvolution and Dynamic Range Improvement
Abstract
dc:description.abstractThe discovery of cells has driven life sciences forward, yet large-field high-temporal-resolution fluorescence microscopy remains hampered by massive data volumes, short exposures and low bit depths that cause photon starvation, motion blur and limited dynamic range. We introduce two complementary computational solutions. First, PEARL deconvolution uses third-order gradient extrapolation to predict and apply optimal update steps, significantly reduce the iteration counts and reconstruction times in both simulated and real 3D datasets—from organelles to organoids—without loss of spatial resolution. Second, Self-Supervised Fluorescence Microscopy High Dynamic Range Reconstruction (Self-FM-HDR) fuses multiple low dynamic range frames with artifact-mitigation strategies to recover a 16-bit-equivalent dynamic range, preserving fine structure in both dim and bright regions under photon-limited conditions. Together, PEARL’s speed and Self-FM-HDR’s expanded intensity range relieve the dual bottlenecks of slow reconstruction rate and low image quality of large-field high-temporal-resolution fluorescence microscopy, possibly laying a robust foundation for next-generation, fast volumetric monitoring of dynamic biological processes.
Degree
thesis:*- Level thesis:degree_level
- Doctoral
- Department dc:contributor.department
- Biomedical Engineering (Joint GT/Emory Department)
- Grantor dc:publisher
- Georgia Institute of Technology
- Year dc:date.issued
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Wu, Shunan
- Advisors dc:contributor.advisor
-
- Wei, Xunbin
- Jia, Shu
- Committee members dc:contributor.committeemember
-
- Xi, Peng
- Li, Changhui
- Luo, Jiajia
Subjects
dc:subject × 5Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/82034