Research and Development of a Novel Phonic Diagnostic System with AI Technologies

2020 Fiscal Year Final Research Report

• Project/Area Number: 18K18407
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 90130:Medical systems-related
• Research Institution: Tenri Health Care University
• Principal Investigator: KOJIMA TSUYOSHI 天理医療大学, 医療学部, 特別研究員 (60767698)
• Project Period (FY): 2018-04-01 – 2021-03-31
• Keywords: 音声障害 / 人工知能 / アプリ / スマートデバイス

Outline of Final Research Achievements

We researched to determine the GRBAS scale, classifying voice disorders into four levels based on auditory impressions, using artificial intelligence (AI). Based on this research, we conducted a study to identify characteristic voice disorders (vocal fold paralysis, supraglottic stenosis, spasmodic dysphonia, etc.) using AI. We have also created and released an iPhone application that uses AI to determine the GRBAS scale for easy use in actual clinical practice. As more data becomes available, it will be possible to create a higher-performance AI, so we have started to consolidate voice and medical data at several collaborating research institutes after standardizing the voice recordings and recording methods.

Free Research Field

耳鼻咽喉科、音声

Academic Significance and Societal Importance of the Research Achievements

本研究ではAIで音声の質を評価できることを示したが、これは熟練した医師や言語聴覚士のようにもしくはそれ以上に音声から診断及び障害の客観的・定量的な評価ができる可能性を示す学術的意義があるものであった。またAIを用いた研究成果をスマートデバイスで使用可能とし、実際に臨床でAIによる音声障害の評価・診断支援を手軽に活かしていくためのための道筋をつけ、検診や救急、地域医療等への幅広い応用の可能性がひろがる社会的意義のある研究となった。