Visual Image Processing for Irregular Phosphene Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current visual prostheses and image processing systems face challenges with irregular phosphene patterns due to limited spatial resolution, dynamic range, and unpredictability, requiring complex calibration processes and ongoing adjustments, which are also applicable to other irregular output image systems like VR and AR.

Innovation Solution

A visual image processing system that decouples sensor maps from visual percepts, allowing flexible and computationally efficient processing by using digital image sensors, depth sensors, or accelerometers to generate stimulus control information based on predefined regions and threshold values, enabling adaptable mapping between stimuli and visual percepts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a training/calibration process is used to map stimuli to phosphenes, then the accuracy of phosphene mapping is improved, but the time required for system setup and ongoing adjustments increases

Engineering Contradiction:
Improvephosphene mapping accuracyVSAvoidcalibration time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary mapping of sensor data regions to phosphene patterns during system setup, creating a lookup table that stores pre-determined stimulus configurations. During operation, the system directly queries this pre-computed mapping rather than performing real-time calibration, significantly reducing operational time while maintaining mapping accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates a simplified computational model that replicates the complex stimulus-phosphene relationships in a condensed format. This model copy allows the system to quickly determine appropriate stimuli without repeatedly performing full calibration procedures, reducing time loss while preserving mapping fidelity.

Inventive Principle:
Principle #26Copying

2Adaptability or versatility

If complex processing is used to handle irregular phosphene patterns, then the adaptability of the system is improved, but the computational complexity increases

Engineering Contradiction:
Improveirregular pattern handling capabilityVSAvoidprocessing complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system divides the visual field into discrete sensor regions, each independently mapped to specific phosphene patterns. This segmentation allows irregular patterns to be handled as composed elements rather than requiring complex holistic processing, reducing computational complexity while maintaining adaptability to irregular configurations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transforms the representation of visual data by changing parameters from continuous irregular coordinates to discrete region-based indexing. This parameter transformation simplifies the mathematical operations required to handle irregular phosphene patterns, reducing processing complexity while preserving the ability to adapt to various pattern configurations.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If iterative refinement with multiple measurement cycles is used, then the phosphene map accuracy is improved, but the productivity of the system decreases

Engineering Contradiction:
Improvephosphene map accuracyVSAvoidsystem efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system performs comprehensive mapping measurements during an initial setup phase, storing the results in a pre-computed lookup table. This preliminary action consolidates multiple measurement cycles into a single setup period, improving phosphene map accuracy while preventing time loss during subsequent operational cycles.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system anticipates future measurement needs by pre-computing and storing multiple stimulus-phosphene mappings in advance. This cushioning approach ensures that even if iterative refinement is needed, the computational work has already been performed, protecting productivity during actual operation while maintaining measurement precision.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentEP3122235B1Image processing system for irregular output patterns
Publication Date: 2020.07.01 MONASH UNIV
  • EP3122235B1 patent drawingFigure 1~2
  • EP3122235B1 patent drawingFigure 3
  • EP3122235B1 patent drawingFigure 4

AI summary

A visual image processing method is based on received spatial field information (302) from a spatial field sensor (116). A data store (804) is accessed, which contains a sensor map data structure (206) comprising a set of predefined regions (208) within a spatial field corresponding with the information received via the sensor input. Each predefined region is associated in the data structure with one or more of a set of stimuli (204) applicable to a biological visual system, and each stimulus corresponds with a visual percept (210). The spatial field information associated with each region is processed to generate stimulus control information which is applied to select, from within the sensor map data structure, stimuli from the set of stimuli for application to the biological visual system. Output stimulus signals (310) are generated, which are suitable for application to the biological visual system based upon the selected stimuli. Flexible mappings are thus provided between visual percepts, stimuli which may be applied (e.g. via a prosthetic implant) in order to generate the percepts, and associations between those stimuli and regions of the spatial field corresponding with the visual percepts.