Ophthalmic Image Enhancement Using Preceding-Frame Tone Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional imaging systems during surgical procedures, such as cataract surgery, face challenges in processing and enhancing image frames quickly enough for real-time display to medical practitioners, leading to sub-optimal surgical outcomes due to lag in capturing, processing, and enhancing current frames.

Innovation Solution

Utilizing enhancement parameters from preceding frames to enhance subsequent frames in the image stream, enabling real-time or near real-time processing by dividing frames into blocks, calculating first-order statistics, and applying tone mapping functions to generate enhanced frames.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If frames are processed in real-time to maintain surgical workflow, then processing speed is improved, but image quality and detail may be compromised

Engineering Contradiction:
Improveprocessing speedVSAvoidimage quality
Core Design Contradiction:
SpeedVSManufacturing precision

Solution Approach 1:

The system performs preliminary calculations of first order statistics (mean, standard deviation) for each block of the current frame before completing full image enhancement. These pre-calculated statistics are then used to generate tone mapping functions for subsequent frames, enabling faster processing while maintaining quality through advance preparation of enhancement parameters.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The image processing is divided into blocks, with first order statistics calculated for individual blocks rather than the entire frame. This segmentation allows parallel processing of different regions and enables selective enhancement of critical areas, improving overall processing speed while preserving image quality through localized optimization.

Inventive Principle:
Principle #1Segmentation

2Manufacturing precision

If enhanced processing is applied to current frames to improve image quality, then manufacturing precision is improved, but processing time increases causing lag

Engineering Contradiction:
Improveimage qualityVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The system calculates first order statistics for the current frame in advance, before the frame is fully processed and displayed. These pre-computed statistical parameters are then reused to generate tone mapping functions for subsequent frames, eliminating redundant calculations and reducing processing time while maintaining enhancement quality.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system copies and reuses first order statistics from the current frame to generate enhancement parameters for subsequent frames. Instead of recalculating statistics for each frame, the system creates tone mapping functions based on copied statistical data, significantly reducing processing time while maintaining consistent image quality across frames.

Inventive Principle:
Principle #26Copying

3Measurement precision

If detailed processing of each frame is performed to capture critical surgical details, then measurement precision is improved, but productivity decreases due to slower frame processing

Engineering Contradiction:
Improvedetail detectionVSAvoidframe processing throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system segments the image into blocks and calculates first order statistics for each block independently. This allows parallel processing of multiple blocks simultaneously, maintaining detailed analysis of critical regions while improving overall frame processing throughput through efficient utilization of processing resources.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes the processing approach by calculating first order statistics (mean, standard deviation) for blocks rather than performing exhaustive detailed processing on entire frames. This parameter-based approach maintains sufficient detail for surgical decision-making while significantly reducing computational complexity and improving processing throughput.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4388490B1Systems and methods for generating enhanced opthalmic images
Publication Date: 2025.09.03 ALCON INC
  • EP4388490B1 patent drawingFigure 1
  • EP4388490B1 patent drawingFigure 2A
  • EP4388490B1 patent drawingFigure 2B

AI summary

Certain aspects of the present disclosure provide an ophthalmic imaging device for enhancing images. The device comprises an image capture component configured to generate an image stream comprising a first frame preceding a second frame, both capturing a branch of veins in an eye. The device further comprises an image processor configured to calculate first order statistics for a plurality of blocks for the first frame and to interpolate first order statistics for the first frame based at least in part on the branch of veins. The image processor is also configured to generate a tone mapping function and calculate tone mapping values for individual pixels of the second frame based on the tone mapping function. The image processor is additionally configured to generate an enhanced frame based on the calculated tone mapping values for individual pixels of the second frame to the pixels of the second frame.