Biometric Apparatus Uniform Image Reconstruction via Iterative Coefficients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Diffuse optical tomography (DOT) measurements result in non-uniform images due to varying spatial resolution and noise characteristics across the measurement region, with regions closer to the surface having better resolution and noise compared to those farther from the surface.
Innovation Solution
A bioinstrumentation apparatus and image generating method that calculates J coefficients for each pixel in the reconstructed image using iterative formulas to ensure convergence rates are uniform across the measurement region, using radiation or sonic waves, and adjusts coefficients to match the slowest convergence rate in each partial region, thereby generating a more uniform image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If diffuse optical tomography is used to measure a living body, then internal information can be obtained non-invasively, but spatial resolution and noise characteristics vary depending on position inside the measurement region resulting in non-uniform images
Solution Approach 1:
The patent applies local quality by dividing the measurement region into multiple partial regions and assigning different regularization coefficients to each region based on its specific characteristics. Regions closer to the surface receive different coefficients than regions farther from the surface, allowing each local area to be optimized for its specific photon flight path characteristics and information content.
Solution Approach 2:
The patent changes the regularization parameter (coefficient) based on the partial region being reconstructed. By adjusting this parameter according to the convergence rate characteristics of each partial region, the system achieves uniform convergence across the entire measurement region, thereby producing uniform images with consistent spatial resolution and noise characteristics throughout.
2Productivity
If successive approximation computation is used for image reconstruction, then internal information can be obtained, but convergence rates vary across different partial regions leading to non-uniform images
Solution Approach 1:
The patent segments the measurement region into multiple partial regions and performs successive approximation computation separately for each region. By dividing the reconstruction process into regional segments, the system can apply region-specific regularization coefficients that account for differences in photon flight paths and information content, achieving uniform convergence across all regions.
Solution Approach 2:
The patent dynamically adjusts the regularization parameter for each partial region based on its convergence rate characteristics. Regions with slower convergence rates receive coefficients that accelerate convergence, while regions with faster convergence rates use coefficients that maintain stability, resulting in uniform convergence behavior across the entire measurement region.
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
The method effectively suppresses differences in spatial resolution and noise characteristics across the measurement region, resulting in a more uniform image quality by adjusting convergence rates and coefficients to match the slowest convergence rate in each partial region.
Implementation Method 1
light which is propagated while being scattered in an interior of the region is detected
Implementation Method 2
an apparatus which uses what-is-called diffuse optical tomography (DOT)
Implementation Method 3
information on a light absorbing body such as a tumor etc. present in the interior of the region can be obtained from measured results of intensity
Data Source
Figure 1
Figure 2
Figure 3(a)~3(b)
AI summary
The bioinstrumentation apparatus 10 includes a light irradiation unit irradiating a measurement region B with light, a light detection unit detecting diffused light from the measurement region, and a computing unit 14 generating a reconstructed image for the interior of the measurement region. The computing unit 14 calculates J coefficients wj set for every pixel of the reconstructed image and more than 0 and not more than 1 (where J is the number of pixels of the reconstructed image) and carries out successive approximation computation by the following iterative formula xjk+1=xjk+wjdjk (where k is an integer from 1 to N, N is the number of times of iterative computation, xj(k) is a pixel value of the jth pixel on the kth iterative computation, and dj(k) is an update amount of the jth pixel on the kth iterative computation) to generate the reconstructed image. Thereby, there are provided a bioinstrumentation apparatus and an image generating method capable of suppressing a difference in spatial resolution and noise characteristics depending on a position inside the measurement region to generate an image which is uniform to a greater extent.