AI Insole Manufacturing with Pressure-Based 3D Interpolation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for designing and manufacturing custom insoles are difficult to automate, particularly due to the complexity of adapting to user anatomy, and current automation techniques involve long processing times and are not user-friendly.

Innovation Solution

A method using an interpolating surface and anchor coordinates for 3D characterization, incorporating computer vision and AI algorithms, allows for automated insole design and production, enabling efficient and fast design adjustments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual annotation of pressure measurements with curves is used, then expert control and customization are improved, but automation capability deteriorates

Engineering Contradiction:
Improveexpert controlVSAvoidautomation capability
Core Design Contradiction:
Ease of operationVSExtent of automation

Solution Approach 1:

The patent replaces manual mechanical annotation processes with automated computer vision and machine learning systems. The system automatically processes pressure measurements and generates insole designs without requiring expert manual curve drawing, thereby maintaining design quality while enabling full automation.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service automation where the computer vision algorithms and machine learning models automatically perform tasks previously requiring expert intervention. The system serves itself by autonomously converting pressure measurements into actionable design parameters without continuous human input.

Inventive Principle:
Principle #25Self-service

2Extent of automation

If brute force application of computer techniques is used, then automation is improved, but processing time increases

Engineering Contradiction:
ImproveautomationVSAvoidprocessing time
Core Design Contradiction:
Extent of automationVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-training machine learning models and establishing decision frameworks before actual insole design. The system pre-processes pressure measurement data into standardized formats and pre-loads material databases, so that when design is needed, the automated system can quickly generate results without extensive real-time computation.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes parameters by transforming raw pressure measurement data into standardized design parameters through machine learning. This parameter transformation enables the system to work with optimized data structures that reduce computational complexity and accelerate processing while maintaining automation.

Inventive Principle:
Principle #35Parameter changes

3Extent of automation

If black box application of computer techniques is used, then automation is improved, but interpretability for operators deteriorates

Engineering Contradiction:
ImproveautomationVSAvoidinterpretability
Core Design Contradiction:
Extent of automationVSLoss of information

Solution Approach 1:

The patent implements feedback mechanisms where the system provides intermediate results and decision rationales to operators during the design process. The machine learning model outputs are accompanied by explanatory information that helps operators understand the automated decisions, maintaining interpretability while preserving automation benefits.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system introduces an intermediary layer between the black box algorithms and the operator. This intermediary presents processed information in an interpretable format, translating complex algorithmic outputs into understandable design recommendations that operators can review and adjust if needed.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4602972A1Fully automatic ai-based manufacturing of an insole
Publication Date: 2025.08.20 ORTHOPEDIE KESTELYN
  • EP4602972A1 patent drawingFigure 1A~2A
  • EP4602972A1 patent drawingFigure 2B~3A
  • EP4602972A1 patent drawingFigure 3B

AI summary

The present invention provides, inter alia, a method for designing an insole, comprising receiving a measurement of a foot sole associated with pressure, said measurement preferably being obtained by placing a load on a measurement surface; receiving a sole profile; determining, based on said measurement, one or more anatomical zones; determining, based on said anatomical zones, a foot profile comprising a major axis; determining, at least based on said foot profile and said anatomical zones, an interpolating surface; aligning the interpolating surface with the sole profile, to obtain an aligned 3D profile; generating, based on said aligned 3D profile, an instruction for manufacturing the insole; wherein said determining of the interpolating surface is realized according to a 3D characterization according to respective positions (y1, y2) on said major axis; wherein said determining of the interpolating surface comprises calculating, based on said foot profile and said anatomical zones, a plurality of at least two anchor coordinates (d0-d3) each lying in a respective cross-section corresponding to said respective positions (y1, y2), and calculating, based on said at least two anchor coordinates (d0-d3), an interpolating surface.