Aphasia Classification via Speech Quantifying and Comprehension Scores
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current aphasia assessments require trained professionals, leading to potential inconsistencies and backlogs, and are limited in accessibility, especially in remote locations with less specialized staff.
Innovation Solution
A machine-learning based system that uses classifiers to analyze spontaneous speech recordings, generating scores and feature values to determine aphasia classifications, which can be administered by laypersons, reducing the need for highly trained staff and standardizing assessments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If trained professionals administer aphasia assessments, then diagnostic accuracy is improved, but accessibility and availability are worsened due to limited specialized staff in remote locations
Solution Approach 1:
The system enables self-service assessment where the automated machine learning model performs the diagnostic evaluation without requiring trained professionals to administer or interpret the tests. The computer system automatically processes speech recordings, extracts features, and generates aphasia classifications, allowing any personnel to conduct assessments while maintaining diagnostic accuracy through the trained ML models.
Solution Approach 2:
The patent replaces the mechanical system of human professional assessment with an automated computer-based machine learning system. The ML models process speech data and generate diagnostic classifications, substituting the need for trained professionals while maintaining or improving diagnostic consistency and accessibility.
2Reliability
If trained professionals administer aphasia assessments, then diagnostic reliability is improved, but time efficiency and productivity are worsened due to assessment backlogs
Solution Approach 1:
The automated system performs diagnostic evaluations independently without requiring professional time for administration or interpretation. The computer system processes multiple assessments simultaneously, eliminating bottlenecks and backlogs while maintaining reliable diagnostic outcomes through the trained machine learning models.
Solution Approach 2:
The system enables continuous, periodic assessment processing where the computer system can handle multiple evaluations in parallel or sequence without fatigue or breaks. This increases throughput and eliminates the periodic constraints of human professional availability, reducing backlogs while maintaining diagnostic reliability.
3Measurement precision
If traditional assessment methods are used, then diagnostic thoroughness is improved, but consistency and standardization are worsened due to potential inconsistencies between different professionals
Solution Approach 1:
The system applies the same machine learning model and processing pipeline uniformly to all assessments, ensuring homogeneous treatment of all patients regardless of location or administering personnel. This eliminates inter-professional variability and ensures consistent, standardized diagnostic criteria are applied universally, improving assessment consistency while maintaining thoroughness through the trained ML models.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing aphasia assessment. One of the methods includes receiving a recording, generating a text transcript of the recording, and generating speech quantifying and comprehension scores which can be used to determine an aphasia classification. Another method includes performing an aphasia assessment on a brain image to obtain an aphasia classification.