Dynamic Morpho-Volumetric Analysis for Lung Lesion Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for diagnosing lung lesions, particularly small-sized lesions, face challenges in reliability and invasiveness, with existing image processing techniques often requiring complex training, high computational costs, and inadequate sensitivity/specificity, especially when dealing with low-sized lesions and physiological evolution of cancer cells.

Innovation Solution

A computer-implemented method using a pipeline that processes moving pictures from CT scans, analyzing geometric and dynamic morpho-volumetric parameters of lung lesions, employing artificial neural networks to classify lesions as benign or malignant, reducing the need for invasive biopsies and incorporating histological characteristics for improved accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If invasive biopsy is performed to obtain reliable diagnosis, then diagnostic reliability is improved, but patient harm and invasiveness increase

Engineering Contradiction:

Improvediagnostic reliabilityVSAvoidinvasiveness

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent replaces the mechanical invasive biopsy procedure with a non-invasive image processing system that uses CT scan images and machine learning algorithms to diagnose lung lesions, thereby eliminating the harmful effects of invasive procedures while maintaining diagnostic reliability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces an intermediary image processing system that acts as a mediator between non-invasive CT imaging and diagnostic decision-making, using extracted features and neural networks to provide reliable diagnosis without direct tissue sampling

Inventive Principle:

Principle #24Intermediary (Mediator)

2Difficulty of detecting and measuring

If classical CT image processing methods are used to differentiate malignant lesions, then diagnostic capability is improved, but sensitivity and specificity remain insufficient especially for small lesions

Engineering Contradiction:

Improvelesion detection capabilityVSAvoidsensitivity and specificity

Core Design Contradiction:

Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The patent transforms the diagnostic approach by changing from analyzing static image features to analyzing dynamic temporal evolution of lesion parameters across multiple CT slices, extracting velocity and acceleration of morpho-volumetric changes which provides superior sensitivity and specificity particularly for small lesions

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent adds a temporal dimension to traditional CT image analysis by examining the evolution of lesion characteristics across multiple time points or slices, transforming static 3D spatial analysis into dynamic 4D spatiotemporal analysis that reveals malignant transformation patterns

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Reliability

If deep learning or machine learning algorithms are applied to classify lung lesions, then classification accuracy is improved, but computational cost and training complexity increase

Engineering Contradiction:

Improveclassification accuracyVSAvoidtraining complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts only the most relevant features from CT images, specifically focusing on temporal evolution of morpho-volumetric parameters rather than processing entire images or all possible features, thereby reducing computational complexity while maintaining high classification accuracy

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments the diagnostic task into distinct stages: extracting temporal evolution features from CT slices, computing morpho-volumetric parameters and their derivatives, and applying neural network classification, allowing each stage to be optimized independently and reducing overall system complexity

Inventive Principle:

Principle #1Segmentation

4Duration of action of stationary object

If follow-up analysis of lung lesions is performed using image registration techniques, then longitudinal assessment is improved, but image registration issues reduce reliability

Engineering Contradiction:

Improvefollow-up capabilityVSAvoidregistration reliability

Core Design Contradiction:

Duration of action of stationary objectVSReliability

Solution Approach 1:

The patent makes the analysis self-referential by computing temporal evolution within each patient's own lesion across their CT slices, eliminating the need to register and compare images across different patients or time points, thereby avoiding registration errors entirely while enabling longitudinal assessment

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11164315B2Image processing method and corresponding system

Publication Date: 2021.11.02 STMICROELECTRONICS SRL

AI summary

A method includes receiving a time series of slice images of medical imaging. The images have a region of interest located at a lung lesion. The method also includes tracking over at least one subset of slice images in a time series of slice images variations over time of at least one image parameter at the set of points in the region of interest. Classifier processing is applied to set of signals indicative of tracked time variations of the at least one image parameter at respective points in the set of points. A classification signal is indicative of the tracked time variations of the at least one image parameter reaching or failing to reach at least one classification threshold.