Biosensor Optical Detection via Spatial Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current biosensing platforms using dark-field micro-spectrophotometry struggle to distinguish different types of nanoparticles and extract fundamental spectral properties, leading to limitations in detecting biomarkers due to averaged optical signals and integral mechanical measurements.
Innovation Solution
A method for optical detection that provides spatially and spectrally resolved optical signals from a biosensor, allowing for parallel analysis of these signals to determine biomarker presence and concentration, with image acquisition and analysis optimized for robustness and speed through particle localization, characterization, classification, and counting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If standard dark-field microscopy is used to detect biomarkers, then the detection can be performed with simple equipment, but the optical signal is averaged over all surface area making it impossible to distinguish different types of nanoparticles
Solution Approach 1:
The patent divides the biosensor surface into multiple individually addressable regions, each containing nanoparticles of a specific type. This spatial segmentation allows different nanoparticle types to be detected simultaneously without signal averaging, enabling both simple equipment operation and precise nanoparticle differentiation.
Solution Approach 2:
The patent introduces a spatial dimension by arranging nanoparticles in specific geometric patterns (triangles, squares, circles) on the biosensor surface. This dimensional organization allows optical detection to distinguish nanoparticle types based on their spatial arrangement rather than relying on complex spectral analysis, resolving the contradiction between equipment simplicity and measurement precision.
2Ease of operation
If standard dark-field microscopy is used for optical recognition, then the measurement process is simple, but fundamental spectral properties of nanoparticles cannot be extracted
Solution Approach 1:
The patent assigns different nanoparticle types to specific local regions on the biosensor surface, each region having unique optical properties. By detecting the optical response at these localized positions rather than averaging over the entire surface, the system maintains operational simplicity while extracting spectral properties of individual nanoparticle types through their spatially separated signals.
3Device complexity
If mechanical transduction is used to detect nanoparticle mass, then the detection mechanism is straightforward, but no information about individual nanoparticles can be obtained
Solution Approach 1:
The patent combines mechanical transduction with spatial segmentation by assigning different nanoparticle types to distinct regions on the biosensor. This allows the mechanical detection system to measure mass changes at specific locations, enabling identification of individual nanoparticle types through their spatial distribution rather than requiring complex individual nanoparticle analysis.
4Measurement precision
If image analysis is performed sequentially after acquisition, then the analysis can be thorough, but the time between image acquisition and results is long
Solution Approach 1:
The patent performs preliminary processing of image data during the acquisition process itself, preparing and analyzing images in parallel as they are captured. This preliminary action eliminates the sequential bottleneck, allowing thorough analysis to be performed without increasing total measurement time, thus resolving the contradiction between analysis thoroughness and measurement speed.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for optically detecting biomarkers in a biosensor, comprising: - simultaneously acquiring (1100) spatially and spectrally resolved images from at least one sample of the biosensor and performing an image analysis (1000) in parallel to the image acquisition (1100); wherein the image analysis (1000) comprises: - reading (2100) data of the acquired images; - correcting (2200) the data to reduce inhomogeneities and noise of the images; - localizing (2300) particles in the images using the corrected data; - characterizing (2400) each particle individually to obtain at least its position and characterization parameters; - classifying (2500) the particles based on their characterization parameters to obtain particle classes; - counting (2600) the particles for each class and acquired image; - for each biomarker in each sample, calculating an overall analysis result (2800) comprising calculating at least one statistical value by using the number of particles per class for all the images acquired from the same sample, and the statistical value per sample being correlated with the presence of a biomarker in the sample.