Passive Acoustic Mapping With Compressive Sensing for Real-Time Cavitation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Real-time processing of growing data streams in passive acoustic mapping (PAM) for therapeutic ultrasound presents a challenge, particularly in monitoring cavitation extent and dose, which is crucial for clinical adoption of high-intensity focused ultrasound (HIFU) and ultrasound-enhanced drug delivery.
Innovation Solution
The implementation of compressive sensing techniques allows PAM to operate directly on sparsely sampled data, using a system with sensor arrays and processing means that define sample periods, random projections, and dictionaries to derive bubble positions from sample data, improving image quality and processing speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional PAM processes all sampled data, then measurement precision is maintained, but processing time increases and real-time capability is lost
Solution Approach 1:
The patent extracts and processes only the most relevant signal components for bubble localization. By identifying and focusing computational resources on significant acoustic events rather than processing all sampled data points, the system maintains measurement precision while reducing processing time to enable real-time monitoring.
Solution Approach 2:
The patent applies partial action by processing a subset of data that contains sufficient information for accurate bubble localization. Rather than exhaustively processing all sampled data, the system identifies key temporal and spatial features that provide the necessary measurement precision with reduced computational burden.
2Quantity of substance
If high channel-count ultrasound platforms are used, then data capture ability improves, but data processing complexity increases
Solution Approach 1:
The patent extracts essential information from the high-volume data stream generated by high channel-count platforms. By identifying and extracting only the critical acoustic features related to cavitation bubble activity, the system maintains comprehensive data capture capability while reducing processing complexity to manageable levels.
Solution Approach 2:
The patent segments the large data stream from high channel-count platforms into manageable components for processing. By dividing the data into distinct temporal and spatial segments that can be independently analyzed, the system preserves the full data capture capacity while reducing overall processing complexity through modular computation.
3Productivity
If sparse sampling is applied, then data processing speed improves, but measurement precision may deteriorate
Solution Approach 1:
The patent performs preliminary identification of significant acoustic events before detailed analysis. By pre-processing the sparsely sampled data to identify potential bubble locations and temporal patterns, the system enables faster processing while maintaining measurement precision through targeted subsequent analysis of only the most relevant data points.
Data Source
AI summary
A passive compression wave imaging system for locating cavitation bubbles, the system comprising a plurality of sensor elements (202b) arranged in an array and each arranged to produce an output signal, and processing means (204a, 204b) arranged to: define a sample period, and a sample space over which the signal can be sampled at each of a plurality of sample points in the sample space; define a grid of candidate bubble positions; define a random projection identifying a respective different group of the sample points for each of the output signals; sample each of the output signals at the group of sample points defined by the random projection to generate sample data; define a dictionary which defines the sample values that would be obtained for the signal of each of the respective group of sensor elements at each of the sample points for the sample period for a bubble at each of the candidate bubble positions; define a vector the elements of which identify the candidate bubble positions; and perform a minimisation operation to derive the vector element values, and hence the bubble positions, from the sample data, the basis and the random projection.