Machine Learning Spinal Analysis from Radiographic Images

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Chiropractors face inefficiencies and inaccuracies in analyzing spinal subluxations from x-ray images due to limited digital analysis tools, leading to missed subtle segmental vertebral hyper-extension or hyper-flexion, which can result in inadequate patient care and treatment outcomes.

Innovation Solution

The implementation of machine learning algorithms for computer-assisted anatomical prediction, allowing for the digitization of anatomical landmarks, determination of linear and angular dimensions, and prediction of anatomical measurements from radiographic images, enabling enhanced analysis and comparison of spinal alignment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional manual measurement methods (protractors, templates) are used for spinal subluxation analysis, then measurement capability is achieved, but time consumption increases and measurement precision decreases

Engineering Contradiction:

Improvespinal subluxation measurement precisionVSAvoidtime for analyzing radiographic images

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical measurement tools (protractors, physical templates) with an automated computer-based system that uses machine learning algorithms to detect anatomical landmarks and calculate spinal measurements automatically, eliminating the need for manual drawing and measurement while improving both speed and precision

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs self-analysis by automatically detecting anatomical landmarks, drawing measurement lines, and calculating spinal subluxation parameters without requiring continuous human intervention, allowing the computer to serve itself in completing the measurement task

Inventive Principle:

Principle #25Self-service

2Measurement precision

If comprehensive segmental and global analysis of spinal alignment is performed, then measurement precision improves, but device complexity and time consumption increase

Engineering Contradiction:

Improvespinal alignment analysis accuracyVSAvoidcomplexity of analysis tools

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the complex spinal analysis task into distinct segments: detection of individual anatomical landmarks (vertebral bodies, spinous processes), drawing of specific measurement lines (midline, spinous process lines), and calculation of separate parameters (segmental rotation, global alignment), allowing each component to be processed independently and systematically

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary machine learning model that acts as a bridge between the raw radiographic image and the final measurement results, automatically performing the complex intermediate steps of landmark detection, line drawing, and coordinate extraction that would otherwise require complex manual procedures

Inventive Principle:

Principle #24Intermediary (Mediator)

3Measurement precision

If manual landmark identification and line drawing is performed, then anatomical measurement capability is achieved, but productivity decreases

Engineering Contradiction:

Improveanatomical landmark identification accuracyVSAvoidpatient care throughput

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the manual mechanical process of identifying landmarks and drawing lines with an automated computer vision system that uses trained machine learning models to detect anatomical features and generate measurement lines automatically, dramatically reducing the time required per patient while maintaining or improving measurement accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs preliminary action by pre-training machine learning models on large datasets of annotated radiographic images before deployment, so that when actual patient images are analyzed, the landmark identification and line drawing operations can be executed rapidly without manual intervention

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11610305B2Method and system for postural analysis and measuring anatomical dimensions from a radiographic image using machine learning

Publication Date: 2023.03.21 POSTURECO INC

AI summary

A method for use of machine learning in computer-assisted anatomical prediction. The method includes identifying with a processor parameters in a plurality of training images to generate a training dataset, the training dataset having data linking the parameters to respective training images, training at least one machine learning algorithm based on the parameters in the training dataset and validating the trained machine learning algorithm, identifying with the processor digitized points on a plurality of anatomical landmarks in a radiographic image of a person's skeleton displayed on a screen by determining anatomical relationships of adjacent bony structures as well as dimensions of at least a portion of a body of the skeleton in the displayed image using the validated machine learning algorithm and a scale factor for the displayed image, and making an anatomical prediction of the person's skeletal alignment based on the determined anatomical dimensions and a known morphological relationship.