Authors:
Hao Zhang
1
;
Guillaume Lopez
1
;
Ran Tao
2
;
Masaki Shuzo
1
;
Jean-Jacques Delaunay
1
and
Ichiro Yamada
1
Affiliations:
1
The University of Tokyo, Japan
;
2
Université de Lyon and INSA-Lyon, France
Keyword(s):
Eating habits monitoring, Sound analysis, Bone-conduction sensors, Wavelet features, Self-Organizing Maps (SOM), Hidden Markov Model (HMM).
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Business Analytics
;
Clinical Problems and Applications
;
Data Engineering
;
Data Mining
;
Databases and Information Systems Integration
;
Datamining
;
Devices
;
Enterprise Information Systems
;
Evaluation and Use of Healthcare IT
;
Health Information Systems
;
Human-Computer Interaction
;
Pattern Recognition and Machine Learning
;
Physiological Computing Systems
;
Sensor Networks
;
Signal Processing
;
Soft Computing
;
Wearable Sensors and Systems
Abstract:
In recent years, an increasing number of people have been suffering from over-weight, indicating the importance
of a balanced dietetic lifestyle. Researches in nutrition and oral health have raised the importance of not
only calorific consumption, but also eating habits quality such as the regularity of meals, eating speed, and
food texture. A new model for the estimation of food texture by analyzing chewing sound collected from a
wearable sensor is presented in this paper. The proposed model combining effective sound features extraction
and classification methods make it possible to estimate quantitatively detailed texture of food a person is eating.
The model has been implemented and shown being efficient (more than 90% accuracy) to estimate three
food texture indices at eight detailed levels for each, with little influence of individual chewing differences.