Machine Learning Surgical Guidance for Anatomy and Step Validation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Surgical mistakes during procedures, particularly in minimally invasive and robotically assisted surgeries, are common due to misidentification of anatomical structures and incorrect procedural steps, exacerbated by limited visibility and lack of haptic feedback.

Innovation Solution

A system utilizing a machine learning medical procedure server and medical procedure systems that generate and apply patient-specific and procedure-specific models to provide guidance and support by analyzing sensor data and providing real-time feedback to prevent errors, using classifiers for step identification, anatomical structure recognition, and tool usage validation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If minimally invasive surgery and robotically assisted surgery are performed, then surgical precision and patient recovery are improved, but the risk of misidentification of anatomical structures and surgical mistakes increases due to limited field of view and lack of haptic feedback

Engineering Contradiction:
Improvesurgical precisionVSAvoidrisk of surgical mistakes
Core Design Contradiction:
Manufacturing precisionVSReliability

Solution Approach 1:

The system implements real-time feedback by continuously monitoring sensor data from the surgical field and providing immediate alerts to the surgeon when anatomical structures are misidentified or when surgical steps are performed out of sequence. The machine learning model analyzes sensor data streams and provides corrective feedback to prevent surgical mistakes.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces an intermediary machine learning system that acts as a mediator between the surgeon and the surgical field. This intermediary processes sensor data, identifies anatomical structures, validates surgical steps, and provides decision support, effectively bridging the gap caused by limited direct visualization and haptic feedback in minimally invasive and robotic surgeries.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If machine learning models are trained on data from multiple medical procedure systems, then model accuracy and adaptability are improved, but data integration complexity and processing requirements increase

Engineering Contradiction:
Improvemodel adaptabilityVSAvoiddata integration complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The machine learning model is designed with universal functionality to process and integrate data from multiple different medical procedure systems and sensor types. The model architecture accommodates diverse data formats and sources, enabling it to learn from heterogeneous datasets across different surgical systems without requiring system-specific customization.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3510506B1Systems and methods for prevention of surgical mistakes
Publication Date: 2021.12.15 VERILY LIFE SCIENCES LLC
  • EP3510506B1 patent drawingFigure 1
  • EP3510506B1 patent drawingFigure 2
  • EP3510506B1 patent drawingFigure 3

AI summary

An apparatus, system and process for guiding a surgeon during a medical procedure to prevent surgical mistakes are described. The system may include a machine learning medical procedure server that generates one or more machine learning medical procedure models using, at least, medical procedure data captured during medical procedures performed at a plurality of different medical procedure systems. The system may also include a medical procedure system communicably coupled with the machine learning medical procedure server that receives a selected machine learning medical procedure model from the machine learning medical procedure server, and utilizes the selected machine learning medical procedure model during a corresponding medical procedure to control one or more operations of the medical procedure system.