Dynamic Image Classification Using Wavelet Time-Frequency Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional methods fail to extract non-stationary signals from dynamic images, particularly those related to peripheral blood vessels, limiting the ability to obtain useful information for diagnosing conditions like pulmonary embolism.

Innovation Solution

A dynamic image classification apparatus and method that performs non-stationary spectrum analysis on dynamic images, utilizing techniques like wavelet transform to analyze time-frequency characteristics and classify images based on these analyses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional stationary spectrum analysis (FFT) is performed on dynamic images, then periodic signals synchronized with cardiac motion can be extracted, but non-stationary signals related to peripheral blood vessels cannot be extracted

Engineering Contradiction:

Improveextraction accuracy of periodic signalsVSAvoidloss of non-stationary signal information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent changes the analysis method from stationary spectrum analysis (FFT) to non-stationary spectrum analysis (wavelet transform), transforming the parameter of signal analysis from frequency-only to time-frequency joint domain, enabling extraction of both periodic and non-stationary signals

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent substitutes the conventional FFT mechanical analysis system with a wavelet transform-based non-stationary spectrum analysis system, replacing the fixed-frequency analysis mechanism with a flexible time-frequency analysis mechanism that adapts to signal characteristics

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Speed

If dynamic imaging is performed with high frame rate to capture detailed motion, then cardiac motion can be resolved, but signal attenuation occurs in lung parenchyma regions

Engineering Contradiction:

Improveframe rateVSAvoidsignal attenuation in lung field

Core Design Contradiction:

SpeedVSLoss of information

Solution Approach 1:

The patent introduces a new dimension of time-frequency analysis by applying wavelet transform to the dynamic image sequences, moving from spatial-temporal domain to time-frequency domain, which enhances the detectability of attenuated signals in lung parenchyma by revealing their frequency characteristics over time

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

3Productivity

If conventional classification methods are used on dynamic images, then processing speed is maintained, but diagnostic accuracy for lung and heart diseases is limited

Engineering Contradiction:

Improveprocessing speedVSAvoiddiagnostic accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent performs non-stationary spectrum analysis as a preliminary processing step before classification, extracting time-frequency features in advance to enhance the input quality for subsequent classification algorithms, thereby improving diagnostic accuracy without significantly increasing overall processing time

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250342584A1Dynamic image classification apparatus, method, and non-transitory computer-readable recording medium storing program

Publication Date: 2025.11.06 KONICA MINOLTA INC

AI summary

A dynamic image classification apparatus according to an embodiment of the present disclosure includes: a hardware processor that receives an input of a result of a non-stationary spectrum analysis on a dynamic image; and a hardware processor that performs classification based on the result of the non-stationary spectrum analysis, which has been inputted.