Physiological Waveform Classification with Validation Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for classifying physiological information waveforms, such as ECG, often misclassify similar segments of waveforms like T waves and QRS complexes due to high similarity, leading to inaccurate recognition.

Innovation Solution

A method involving a computer-based system that acquires waveform data, classifies it into predetermined types, determines the validity of the classification, and corrects the classification results based on validity, using overlapping time durations and machine learning techniques like deep learning to improve accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If narrow segments are used for waveform classification, then processing speed is improved, but classification accuracy deteriorates due to high similarity between different waveforms

Engineering Contradiction:

Improveprocessing speedVSAvoidclassification accuracy

Core Design Contradiction:

SpeedVSMeasurement precision

Solution Approach 1:

The patent applies segmentation by dividing the waveform into multiple narrow segments for parallel processing, which improves processing speed. Each segment is independently classified, allowing simultaneous analysis of multiple waveform portions without sacrificing overall accuracy when combined with the validation and correction mechanisms.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements feedback through the validation and correction process. The classification results are validated against expected waveform characteristics, and corrections are applied when misclassifications are detected. This feedback loop compensates for the accuracy loss from using narrow segments, maintaining high classification accuracy while preserving processing speed benefits.

Inventive Principle:

Principle #23Feedback

2Measurement precision

If validation and correction steps are added to the classification process, then classification accuracy is improved, but processing time increases

Engineering Contradiction:

Improveclassification accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies preliminary action by performing validation checks using expected waveform characteristics before finalizing classification results. This preliminary validation prevents obviously incorrect classifications from being processed further, reducing the need for extensive correction work and minimizing additional processing time while improving accuracy.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes parameters by adjusting the strictness of validation criteria and correction thresholds based on the specific waveform type being analyzed. By dynamically adjusting these parameters, the system optimizes the balance between accuracy improvement and processing time, applying more rigorous validation only when necessary to resolve ambiguous cases.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS11317845B2Physiological information waveform processing method, program, computer readable storage medium, and physiological information waveform processing apparatus

Publication Date: 2022.05.03 NIHON KOHDEN CORP

AI summary

A method is implemented by a computer, and includes: (a) acquiring at least one set of waveform data having a time duration from physiological information waveform data; (b) classifying a waveform included in the waveform data into a predetermined type of waveform; (c) determining validity of a classification result of the waveform; and (d) correcting the classification result in accordance with the validity of the classification result.