Automated Osseous Image Segmentation for Forensic Identification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current forensic identification methods, particularly skeletal forensic identification, rely heavily on manual and subjective comparisons of radiographs, which are time-consuming, prone to errors, and lack objectivity, especially in scenarios with limited or degraded ante-mortem and post-mortem data, such as in mass disaster victim identification.

Innovation Solution

An automated method using artificial intelligence and computer vision techniques for the comparison of osseous images, including deep learning-based segmentation and registration, eliminates the need for manual intervention and allows for the integration of multiple anatomical structures, providing a more objective and reliable decision-support system for forensic identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual comparison of radiographs is used for forensic identification, then expert judgment and experience can be applied, but the process is time-consuming and prone to errors

Engineering Contradiction:
Improveidentification accuracyVSAvoididentification time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical comparison process with an automated computer vision system. The system uses image processing algorithms to automatically compare ante-mortem and post-mortem radiographs, substituting human expert manual inspection with computational analysis. This automation maintains identification accuracy while dramatically reducing the time required for forensic comparison.

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

Solution Approach 2:

The system enables self-service by allowing the comparison process to execute autonomously without continuous human intervention. The automated algorithm independently performs image registration, feature extraction, and similarity assessment, freeing experts from tedious manual work while maintaining reliable identification results.

Inventive Principle:
Principle #25Self-service

2Reliability

If manual comparison methods are used, then flexibility in handling diverse cases is maintained, but objectivity and reproducibility are compromised

Engineering Contradiction:
ImproveobjectivityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces subjective manual assessment with objective computational analysis. The computer vision system applies consistent algorithms to all comparisons, eliminating human bias and variability. This substitution provides measurable, reproducible results that enhance objectivity while the modular system design manages complexity through standardized processing pipelines.

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

3Measurement precision

If comprehensive skeletal structure analysis is performed manually, then detailed examination is possible, but the process becomes excessively lengthy and fatiguing

Engineering Contradiction:
Improvecomparison precisionVSAvoididentification throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent segments the comprehensive skeletal analysis into distinct processing stages: image preprocessing, registration, feature extraction, and comparison. This segmentation allows the system to perform detailed examinations of multiple skeletal structures simultaneously through automated parallel processing, maintaining high precision while dramatically increasing identification throughput compared to sequential manual analysis.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The automated system enables continuous processing of multiple cases without the fatigue that limits manual analysis. The computer vision algorithm can continuously compare radiographs with consistent precision, maintaining high measurement accuracy across large numbers of cases without the declining performance that occurs during prolonged manual examination.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP3905129B1Method for identifying bone images
Publication Date: 2025.03.26 UNIV DE GRANADA
  • EP3905129B1 patent drawingFigure 1~2
  • EP3905129B1 patent drawingFigure 3
  • EP3905129B1 patent drawingFigure 4

AI summary

The present invention has as its objective a procedure for assisting a forensic expert in making decisions in order to identify subjects by comparing images of rigid anatomical structures. This procedure includes a decision-making stage based on a hierarchical analysis model that, in particular realizations, is complemented by a previous stage of segmentation of osseous images and their superimposition.