Concept Graph Neural Networks for Intraoperative Surgical State Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing surgical care systems face challenges in providing equitable access to high-quality surgical procedures, particularly in rural areas and for minority populations, due to regionalization and limited availability of high-volume hospitals, and existing decision-making models are inadequate in addressing cognitive errors and inefficiencies in intraoperative decision-making.

Innovation Solution

A system utilizing concept graph neural networks to analyze intraoperative sensor data, including video and biometric parameters, to generate statistical parameters representing the state of a surgical procedure, enhancing surgical decision-making through automated predictive-assistive tools.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If cognitive task analysis is used to codify experienced surgeons' knowledge into standardized checklists, then decision-making support is improved, but 50-75% of decisions made in surgery can be lacking in conscious recall due to inexperience or automaticity, and these efforts are time consuming

Engineering Contradiction:

Improvedecision-making supportVSAvoidtime consuming

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual cognitive task analysis and checklist-based decision support with an automated machine learning system that processes sensor data, video, and audio to generate decision support parameters in real-time, eliminating the time-consuming manual codification process while maintaining or improving reliability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an intermediary machine learning system that acts as a bridge between raw surgical data and decision support outputs, automatically extracting meaningful parameters without requiring direct human codification of every decision scenario

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If regionalization of surgical care is implemented to concentrate operations at high-volume hospitals, then surgical care quality is improved, but access to surgery is restricted for rural areas and minority populations

Engineering Contradiction:

Improvesurgical care qualityVSAvoidaccess to surgery

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The patent creates a virtual copy of expert surgical knowledge through machine learning models trained on data from high-volume hospitals, allowing this expertise to be distributed and applied at lower-volume facilities through automated decision support systems, thereby maintaining quality while improving access

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent develops a universal machine learning system that can be deployed across multiple surgical facilities with varying volumes and resources, providing standardized decision support functionality that adapts to different local contexts without requiring physical presence at centralized high-volume centers

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Ease of operation

If surgeons learn from one patient at a time through apprenticeship during residency, then fundamental surgical skills are developed, but knowledge on rare procedures is limited due to limited availability of experienced surgeons

Engineering Contradiction:

Improvefundamental surgical skillsVSAvoidknowledge on rare procedures

Core Design Contradiction:

Ease of operationVSLoss of information

Solution Approach 1:

The patent merges data from multiple surgical cases, including rare procedures, into a unified machine learning model that captures patterns and knowledge across diverse scenarios, allowing trainees and practitioners to access aggregated expertise that would be unavailable through individual case experience alone

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent replaces the apprenticeship-based knowledge transfer mechanism with an automated machine learning system that can efficiently process and transmit knowledge about rare procedures from experienced surgeons to broader communities, overcoming the limitations of limited personal exposure

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12362068B2Interpretation of intraoperative sensor data using concept graph neural networks

Publication Date: 2025.07.15 THE GENERAL HOSPITAL CORP

AI summary

Systems and methods are provided for generating a statistical parameter representing a state of a surgical procedure from sensor data. Sensor data representing a time period. is received from a sensor. Numerical features representing the time period are generated from the sensor data. Each of a plurality of long short term memory units are updated according to the plurality of numerical features via a message passing process. The long short term memory units are connected to form a graph, with a first set of the long short term memory units representing a plurality of nodes of the graph and a second set of the long short term memory units representing a plurality of hyperedges of the graph. A statistical parameter representing a state of the surgical procedure for the time period is derived from an output of one of the long short term memory units and provided to a user.