Development of physical activity evaluation method using wearable device by voice input

2019 Fiscal Year Final Research Report

• Project/Area Number: 17K01800
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Applied health science
• Research Institution: Nihon University
• Principal Investigator: NAMBA Hideyuki (難波秀行) 日本大学, 理工学部, 准教授 (80559790)
• Project Period (FY): 2017-04-01 – 2020-03-31
• Keywords: 身体活動 / 音声認識 / 行動記録 / スマートフォン / 加速度計 / METs

Outline of Final Research Achievements

The use of Web-based physical activity systems has been proposed as an easy method for collecting physical activity data.Objective　of this study was to develop a behavior-recording application (APP) using voice recognition. The results from our developed APP were compared with objective data from a 3-axis accelerometer to assess the strengths and weaknesses of the new measurement system. A total of 20 participants (14 men, 6 women, 19.1 (SD 0.9) years of age) wore a 3-axis accelerometer and inputted behavioral data into their smartphones for a period of 7 days. The measure of intensity was metabolic equivalents (METs). The Pearson correlations for the METs between the two methods were all positive and significant when the analysis was for over 10 hours, r = 0.545 (P =.017), and for over 14 hours with voice input, r = 0.750 (P =.008). Voice recognition APP appear to be useful for assessing physical activity with high accuracy. However, voice input compliance is an important factor.

Free Research Field

健康科学，スポーツ科学

Academic Significance and Societal Importance of the Research Achievements

開発したスマートフォンを用いた音声入力による行動記録法は，多人数の身体活動を簡便に精度よく評価することができるため，疫学研究などに応用できる可能性がある．さらに，自身の身体活動量を加速度計など高価な機器を用いなくても評価できるため，身体不活動が原因となる疾患の予防や健康づくりに役立てられる可能性がある．精度よく評価するためには，正しく入力をすることが重要であることが研究結果より示された．