Encoded PSF Microscope for 3D Light Field Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D microscope technologies face limitations such as shallow depth of field, loss of depth position information, inability to deconvolve point spread functions (PSFs) from extended objects, and reduced lateral resolution, making them inefficient for imaging complex or fast-moving biological samples.
Innovation Solution
A microscope system using scanned illumination with a pixelated detector and a specially designed phase plate to capture the Light Field of extended objects, allowing for encoded PSF identification and 3D imaging, enabling accurate depth localization and extended depth of field without the need for multi-focal stacks or pinholes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of stationary object
If confocal or widefield deconvolution microscopes are used to overcome shallow depth of field, then depth of field is improved, but imaging speed decreases and fluorescence bleaching increases
Solution Approach 1:
The patent uses periodic scanning of a single illumination spot across the sample in a raster pattern, illuminating different regions sequentially rather than simultaneously. This periodic scanning approach allows the system to achieve extended depth of field through encoded PSF measurement while maintaining high imaging speed, as each spot position is measured rapidly and the entire sample is covered through systematic scanning.
Solution Approach 2:
The patent extracts only the necessary depth information by measuring encoded PSFs at each scanned spot position rather than capturing complete multi-focal stacks. By taking out only the essential depth-encoded PSF measurements and using orthogonal encoding schemes, the system achieves depth of field extension without the excessive data acquisition requirements of traditional confocal or widefield deconvolution methods.
2Loss of information
If multi-focal stack imaging is used to capture depth information, then depth coverage is improved, but imaging speed decreases and photodamage increases
Solution Approach 1:
The patent changes the parameter of PSF encoding by using orthogonal encoded PSFs with distinct spatial signatures at different depths. Instead of capturing multiple focal stacks, the system modifies the PSF structure through encoding schemes that embed depth information directly in the PSF pattern, allowing single-shot depth measurement and dramatically reducing imaging time while preserving complete depth position information.
Solution Approach 2:
The patent replaces the mechanical multi-focal stacking approach with a computational encoding system. Rather than physically moving the focal plane through mechanical z-stack acquisition, the system uses encoded PSFs that carry depth information inherently, substituting mechanical depth scanning with optical encoding and computational decoding to achieve faster depth information acquisition.
3Measurement precision
If rotating PSFs are used for super-localization in z, then depth localization accuracy is improved, but applicability decreases to only sparse point-like objects
Solution Approach 1:
The patent creates a universal measurement system that works for both sparse point-like objects and extended objects by using orthogonal encoded PSFs. The encoding scheme is designed to be applicable across different object types, allowing the same system to achieve super-localization accuracy for points while also enabling depth mapping of extended structures, thus achieving multi-functionality and broad adaptability.
Solution Approach 2:
The patent segments the extended object into multiple scanned spot positions, measuring encoded PSFs sequentially across the sample. By dividing the extended object measurement into discrete spot measurements and using orthogonal encoding that remains valid across different spatial locations, the system achieves depth localization accuracy comparable to point sources while extending applicability to continuous extended structures.
4Adaptability or versatility
If Light Field photography is used to capture depth information, then post-capture refocusing capability is improved, but lateral resolution decreases significantly
Solution Approach 1:
The patent applies local quality by using a pixelated detector that resolves the exit pupil structure with high spatial frequency, preserving local resolution information. The encoded PSFs are measured with sufficient spatial detail in the pupil plane to maintain lateral resolution while enabling post-capture refocusing. This local quality in the pupil plane measurement ensures that resolution is not sacrificed for the added capability of computational refocusing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables high-speed, accurate 3D imaging and animation of subcellular events by separating and decoding PSFs, allowing for precise depth localization and visualization of complex biological features without the limitations of existing technologies.
Implementation Method 1
capture the entire Light Field of the object over the scan. A pixelated detector (e.g. a 5×5 or 10×10 photodiode array readable at thousands of frames/second) images the exit pupil of the microscope objective
Implementation Method 2
The object is illuminated by a scanned spot (or set of spots) of light. A series of images is generated. Computations generate a Light Field image of the entire 3D object
Data Source
AI summary
Scanned illumination allows for capturing 3-dimensional information about an object. A conventional reflection (or transmission) bright field (or fluorescence, darkfield, polarizing, phase contrast or interference) microscope is configured to use laterally scanned illumination (for example by moving an array in front of the light source) to scan an extended object. A pixelated detector may capture a series of images at the exit pupil of a microscope objective, and this series of images may be processed to form a Light Field image of the object. Or, a microscope is configured to provide scanned illumination to an extended object, while applying extended depth of field and 3D depth localization encoding to the resulting set of images. Thus multiple encoded images are generated. These images are decoded and combined, with custom digital signal processing algorithms, to form a 3D volume rendering or animation. Each point in the specimen is illuminated separately from its neighbors, and records its distinct PSF signature without any ambiguity arising from adjacent points on the object.