Audiogram Classification System Using Segmented Rules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current audiogram classification systems are impractical due to their complexity, lack of standardization, and failure to account for local irregularities, leading to inconsistent interpretations among audiologists, which complicates the categorization and analysis of hearing loss trends.
Innovation Solution
A method for classifying audiograms into standardized categories using a system that selects configuration, severity, site of lesion, and symmetry, with rules that ignore local irregularities and maximize agreement among judges, and an automated classification system that generates hearing aid prescriptions based on these classifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a classification system provides numerous categories, subcategories, labels, subscripts, and superscripts to accommodate numerous interpretations, then the system can cover more audiogram variations, but the system becomes too complex for practical clinical application
Solution Approach 1:
The classification system is divided into multiple hierarchical levels: configuration categories (7 types), severity categories (5 levels), site of lesion categories (4 types), and symmetry categories (2 types). This segmentation allows comprehensive coverage of audiogram variations while keeping each individual category simple and manageable for clinical use.
Solution Approach 2:
The classification system uses a universal framework that can handle diverse audiogram types through a standardized set of categories and rules. The same classification structure accommodates different hearing loss patterns, configurations, and severity levels without requiring separate classification systems for each scenario.
2Ease of operation
If classification systems provide general rules for placing audiograms in categories, then the system is easier to use, but the system does not deal with the practical issue of assigning a category when audiologists disagree or when there are local irregularities
Solution Approach 1:
The system incorporates specific quantitative parameters and thresholds for classification decisions. For example, severity is determined by specific dB HL ranges (mild: 26-40 dB, moderate: 41-60 dB, severe: 61-80 dB, profound: 81-100 dB), and configuration is determined by threshold patterns across frequency ranges. These parameter-based rules reduce subjectivity and improve consistency.
Solution Approach 2:
The system introduces an intermediary layer of standardized rules that mediate between the raw audiogram data and the final classification. These rules act as a bridge that translates variable audiogram presentations into consistent categories, handling local irregularities through predefined criteria rather than direct audiologist judgment.
3Adaptability or versatility
If audiologists manually interpret audiograms using personal and subjective rules, then the interpretation reflects individual expertise, but the categorization is difficult to analyze and compare for studying hearing loss trends
Solution Approach 1:
The system transforms subjective audiogram interpretations into standardized parameter-based categories. Configuration is classified by threshold patterns (flat, sloping, rising, trough, peaked, irregular), severity by dB ranges, site of lesion by air-bone gap patterns, and symmetry by interaural differences. This parameterization enables systematic analysis while preserving the essential characteristics of different hearing loss types.
4Productivity
If an automated classification system is implemented, then the system is easier to administer and provides more consistent results, but the system requires standardized rules that may not account for all clinical nuances
Solution Approach 1:
The automated system processes classification in segmented steps: first determining configuration based on threshold patterns, then severity based on dB levels, then site of lesion based on air-bone gaps, and finally symmetry based on interaural comparisons. This segmented approach simplifies automation while maintaining comprehensive classification.
Solution Approach 2:
The system is designed to be self-sufficient with clearly defined algorithms that automatically determine all classification categories without requiring external expert intervention. The standardized rules enable the system to independently classify audiograms consistently, reducing administrative burden while maintaining reliability.
Data Source
AI summary
An audiogram classification system is provided. The classification system includes categories for configuration, severity, site of lesion and/or symmetry of an audiogram. A set of rules can be provided for selecting the categories, wherein the set of rules ignore one or more local irregularities on an audiogram and have been validated to maximize agreement with judges.