Eating detection and chews counting through sensing mastication muscle contraction

Wang, Shuangquan; Zhou, Gang; Ma, Yongsen; Hu, Lisha; Chen, Zhenyu; Chen, Yiqiang; Zhao, Hongyang; Jung, Woosub

doi:10.1016/j.smhl.2018.07.004

Cited by 27 publications

(13 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to create a model, the LOSO employs samples of all subjects but leaves one out to shape the training data. Then, using the samples of the excluded subject, the trained model is examined [ 51 ].…”

Section: Proposed Methodologymentioning

confidence: 99%

LSTM Networks Using Smartphone Data for Sensor-Based Human Activity Recognition in Smart Homes

Mekruksavanich

Jitpattanakul

2021

Sensors

198

View full text Add to dashboard Cite

Human Activity Recognition (HAR) employing inertial motion data has gained considerable momentum in recent years, both in research and industrial applications. From the abstract perspective, this has been driven by an acceleration in the building of intelligent and smart environments and systems that cover all aspects of human life including healthcare, sports, manufacturing, commerce, etc. Such environments and systems necessitate and subsume activity recognition, aimed at recognizing the actions, characteristics, and goals of one or more individuals from a temporal series of observations streamed from one or more sensors. Due to the reliance of conventional Machine Learning (ML) techniques on handcrafted features in the extraction process, current research suggests that deep-learning approaches are more applicable to automated feature extraction from raw sensor data. In this work, the generic HAR framework for smartphone sensor data is proposed, based on Long Short-Term Memory (LSTM) networks for time-series domains. Four baseline LSTM networks are comparatively studied to analyze the impact of using different kinds of smartphone sensor data. In addition, a hybrid LSTM network called 4-layer CNN-LSTM is proposed to improve recognition performance. The HAR method is evaluated on a public smartphone-based dataset of UCI-HAR through various combinations of sample generation processes (OW and NOW) and validation protocols (10-fold and LOSO cross validation). Moreover, Bayesian optimization techniques are used in this study since they are advantageous for tuning the hyperparameters of each LSTM network. The experimental results indicate that the proposed 4-layer CNN-LSTM network performs well in activity recognition, enhancing the average accuracy by up to 2.24% compared to prior state-of-the-art approaches.

show abstract

Section: Proposed Methodologymentioning

confidence: 99%

LSTM Networks Using Smartphone Data for Sensor-Based Human Activity Recognition in Smart Homes

Mekruksavanich

Jitpattanakul

2021

Sensors

198

View full text Add to dashboard Cite

show abstract

“…A single-axis accelerometer uses to measure temporalis muscle movement. The research was then extended using a triaxial accelerometer and analyze other activities such as standing, walking, sitting reading, and coughing separately [39]. Both of the research embedded the accelerometer into a headband and attach the sensors to the temporalis muscle by using medical tape.…”

Section: ) Accelerometermentioning

confidence: 99%

“…Among the food type that used, watermelon categorized as a unique selection. In [39], researchers are only evaluating watermelon in their research with the reasons of watermelon is solid food that required less effort in chewing it. Hence, if the device could accurately capture the chewing activity, it can be assumed that the device can detect other types of foods that require big effort in chewing.…”

Section: Chewing Datasetmentioning

confidence: 99%

“…An EMG sensor signals were sampled at the range of 1024Hz [31], 44.1kHz and resampled at 1kHz [32], 25 Hz and 50Hz [3], 512Hz [33] and 256Hz [34], [35]. In [39] and [40], the accelerometer signal sampled at 100Hz. While for other sensing devices, the sampling frequency is 20Hz for PPG [41], 10 Hz for optical sensors [42], and 192kHz for the doppler sensor in [43].…”

Section: Chewing Classificationmentioning

confidence: 99%

“…In [23], Sazonov & Fontana normalizing the signal to signal the median. Research by Wang et al in [39] uses z-score normalization to eliminate the scaling effects among different features. While in [58], normalizations used to compensate for the difference in the features unit.…”

Section: A Data Pre-processingmentioning

confidence: 99%

See 2 more Smart Citations

Automatic Food Intake Monitoring Based on Chewing Activity: A Survey

Selamat

Ali

2020

IEEE Access

View full text Add to dashboard Cite

Good nutrition is essential for optimal growth, development, and prevention of disease. Due to the importance of nutrition in human life, researchers have been interested in understanding the science of assessing food intake episodes for decades. With the advancement of technology, automated food monitoring tool develops with the help of sensors to address issues related to self-reporting methods. Food monitoring technology is evolving rapidly due to the advancement of sensors; however, automatic monitoring of food intake remains open problems to be solved. For food intake episode detection and monitoring, the sensors used to detect bites, chew, swallow, and hand gestures movement. This survey will be focusing on chewing activity detection during eating episodes. In this survey, a wide range of chewing activity detection explored to outline the sensing design, classification methods, performances, chewing parameters, chewing data analysis as well as the challenges and limitations associated with them.INDEX TERMS Food intake monitoring, automatic food intake detection, chewing sensor, chewing detection, chewing classification, chewing count, chewing data analysis.

show abstract

Antagonistically Functionalized Diatom Biosilica for Bio‐Triboelectric Generators

Rajabi‐Abhari

Lee

Tabassian

et al. 2022

Small

View full text Add to dashboard Cite

Although biomaterial‐based triboelectric nanogenerators (Bio‐TENGs) for use in wearable electronics and implantable sensors have been developed, power generation is not suitable for satisfying the basic requirements for practical applications. Here, to greatly enhance output performances of Bio‐TENG devices, an antagonistic approach of diatom frustules (DFs) with amine and fluorine chemical functionalizations is reported. The DFs are treated with piranha solution to increase the density of hydroxyl groups and tribo‐positive and tribo‐negative composite films are designed with antagonistically functionalized DFs. The tribo‐positive composites having electron donating functionality consist of aminated DFs and cellulose nanocrystals (CNCs), while the tribo‐negative composite is composed of fluorinated DFs and polydimethylsiloxane (PDMS). An antagonistically and chemically functionalized TENG (ACF TENG) with an efficient contact area of 9.6 cm2 under a force of 8 N and a frequency of 5 Hz exhibits an output voltage of 248 V, a short‐circuit current of 16.4 µA, and a power density of 2.01 W m−2, which is 16.6 times higher than a reference (CNC:PDMS) TENG. This study shows a simple antagonistic approach for chemical functionalization as an efficient method to manipulate the tribo‐polarity of bio‐additives for enhancing power generation of Bio‐TENGs.

show abstract

Eating detection and chews counting through sensing mastication muscle contraction

Cited by 27 publications

References 20 publications

LSTM Networks Using Smartphone Data for Sensor-Based Human Activity Recognition in Smart Homes

LSTM Networks Using Smartphone Data for Sensor-Based Human Activity Recognition in Smart Homes

Automatic Food Intake Monitoring Based on Chewing Activity: A Survey

Antagonistically Functionalized Diatom Biosilica for Bio‐Triboelectric Generators

Contact Info

Product

Resources

About