Intraoperative Pupil Morphology Detection with Obstruction-Aware AI

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for determining pupil morphology during ophthalmic surgical procedures, particularly cataract surgery, face challenges due to surgical illumination variations, instrument obstruction, and lens material hydration, leading to inaccuracies in pupil recognition and instability, which can complicate surgeries.

Innovation Solution

A system utilizing adaptive wavelet tensor feature extraction, anatomy segmentation, and obstruction classification machine learning models processes surgical video data to accurately determine and normalize pupil size, accounting for obstructions and variations, providing real-time notifications to surgeons.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If conventional computer vision-based methods or deep learning models are used for pupil recognition, then automation is improved, but measurement precision deteriorates due to surgical illumination variations, instrument obstruction, and lens material hydration

Engineering Contradiction:

Improveautomated pupil recognitionVSAvoidpupil recognition accuracy

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The patent segments the surgical video data processing into distinct stages: feature extraction from surgical images, segmentation of anatomical structures (pupil, iris, lens), and separate analysis of obstruction elements. This multi-stage segmentation allows each component to be optimized independently, improving overall measurement precision while maintaining automation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces intermediate processing steps between raw surgical images and final pupil measurements, including feature extraction layers, segmentation masks, and obstruction classification. These intermediaries filter and prepare data at each stage, reducing the impact of illumination variations and obstructions on the final measurement accuracy.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Adaptability or versatility

If multiple surgical phases with different instrumentation are captured, then adaptability is improved, but measurement precision deteriorates due to varying appearances of the pupil intraoperatively

Engineering Contradiction:

Improveprocessing multiple surgical phasesVSAvoidpupil size determination accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent implements dynamic adaptation by training the machine learning model on diverse surgical phase data, allowing the system to adjust its feature extraction and segmentation parameters based on the current surgical phase characteristics. This enables accurate pupil measurement across varying instrumental conditions and pupil appearances.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes processing parameters dynamically based on surgical phase detection, adjusting illumination normalization, feature extraction weights, and segmentation thresholds according to the specific phase being captured. This maintains measurement precision across different surgical conditions.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If real-time pupil monitoring is implemented, then productivity is improved, but device complexity increases due to multiple machine learning models and processing stages

Engineering Contradiction:

Improvereal-time pupil monitoringVSAvoidsystem architecture

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent merges multiple machine learning functions (feature extraction, segmentation, obstruction detection, pupil measurement) into an integrated processing pipeline. This consolidation maintains real-time monitoring capability while reducing overall system complexity through unified model architecture and shared computational resources.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent creates a universal processing framework that handles multiple surgical phases, different instrumentation types, and various obstruction scenarios through a single integrated system. This multi-functional approach eliminates the need for separate specialized systems, maintaining productivity while managing complexity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250232436A1Determining changes in pupil morphology during an ophthalmic surgical procedure

Publication Date: 2025.07.17 THE RGT UNIV OF MICHIGAN

AI summary

System and methods for determining pupil morphology during an ophthalmic surgical procedure. The system may receive surgical video data of an ophthalmic surgical procedure. The surgical video data may include images of at least a portion of an eye. The system may provide the surgical video data to an adaptive wavelet tensor feature extraction machine learning (ML) model to generate preprocessed image data, including preprocessed images with extracted features associated with anatomical components of the eye. The system may provide the preprocessed image data to an anatomy segmentation ML model to generate segmented image data and segmentation masks. The system may provide the segmented image data and segmentations masks to an obstruction classifier ML model to determine a size of a pupil of the eye in the segmented images. The system may generate a notification based upon the size of the pupil, and provide the notification to a user device.