Automatic Food Intake Monitoring Based on Chewing Activity: A Survey

Selamat, Nur Asmiza; Ali, Sawal Hamid Md

doi:10.1109/access.2020.2978260

Cited by 27 publications

(7 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For simplicity, we worked with data sampled at 100 Hz in all experiments. However, similar studies [ 22 , 63 ] show that eating could be detected with data sampled at 25 Hz, which is important for practical implementation. Next, the signals were filtered using a fifth-order median filter.…”

Section: Eating Detection Approachmentioning

confidence: 98%

Smartwatch-Based Eating Detection: Data Selection for Machine Learning from Imbalanced Data with Imperfect Labels

Stankoski

Jordan

Gjoreski

et al. 2021

Sensors

View full text Add to dashboard Cite

Understanding people’s eating habits plays a crucial role in interventions promoting a healthy lifestyle. This requires objective measurement of the time at which a meal takes place, the duration of the meal, and what the individual eats. Smartwatches and similar wrist-worn devices are an emerging technology that offers the possibility of practical and real-time eating monitoring in an unobtrusive, accessible, and affordable way. To this end, we present a novel approach for the detection of eating segments with a wrist-worn device and fusion of deep and classical machine learning. It integrates a novel data selection method to create the training dataset, and a method that incorporates knowledge from raw and virtual sensor modalities for training with highly imbalanced datasets. The proposed method was evaluated using data from 12 subjects recorded in the wild, without any restriction about the type of meals that could be consumed, the cutlery used for the meal, or the location where the meal took place. The recordings consist of data from accelerometer and gyroscope sensors. The experiments show that our method for detection of eating segments achieves precision of 0.85, recall of 0.81, and F1-score of 0.82 in a person-independent manner. The results obtained in this study indicate that reliable eating detection using in the wild recorded data is possible with the use of wearable sensors on the wrist.

show abstract

Section: Eating Detection Approachmentioning

confidence: 98%

Smartwatch-Based Eating Detection: Data Selection for Machine Learning from Imbalanced Data with Imperfect Labels

Stankoski

Jordan

Gjoreski

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…In clinical practice, arterial pulse and chewing are prominent components of human physiological parameters and involve abundant health information. [60][61][62][63] The former is the general signal that requires to be frequently monitored, while the latter could be of help to the treatment or nursing of certain specific diseases, such as Alzheimer's disease and high paraplegia. Here, we integrated the proposed hybrid device with a customized eyeglass and utilized the signal collection and wireless transmission of the ESP32 MCU to build a wearable smart eyeglass system for monitoring physiological parameters in real time, including temporal pulse and chewing.…”

Section: Smart Integrated Systems For Tracking Physiological Paramete...mentioning

confidence: 99%

Wearable Hybrid Device Capable of Interactive Perception with Pressure Sensing and Visualization

Guo

et al. 2022

Adv Funct Materials

View full text Add to dashboard Cite

A wearable interactive device that can synchronously detect and visualize pressure, stimuli will supplement easily identifiable in situ signals with conventional sensing that outputs only electronic signals. Currently, most interactive devices are composed of physically connected separate pressure sensors and display elements, which do not conform to the development of miniaturization and high integration. The realization of combining these two functions in one device with a simplified configuration is still only beginning to be explored. Inspired by the two-in-one response of the octopus, which can both sense and visualize stimuli, an integrated hybrid device based on ionic sensing and electrochromic display is proposed for interactive pressure perception. By efficiently reusing the electrodes and ionic gel, the device enables quantitative sensing and direct color changing in response to pressure. Benefiting from its excellent comprehensive performance, the applications are explored and verified, including the interactive perception of pressure information, hiding and display of visual information under pressure regulation, and repeated writing and erasing on flexible boards. Notably, multiple customized systems with improved design flexibility are implemented to provide remote monitoring and visual warning of behavioral states and physiological parameters. This study diversifies solutions to realize a direct link between pressure detection and visualization.

show abstract

“…There has been promising work on technology-driven methodologies broadly as summarized in 19,[59][60][61][62] . Many have taken a wearable sensor approach typically developed for individual use (in contrast to agents monitoring intake of others).…”

Section: Related Workmentioning

confidence: 99%

Enhancing Food Intake Tracking in Long-Term Care with Automated Food Imaging and Nutrient Intake Tracking (AFINI-T) Technology

Pfisterer¹,

Amelard²,

Boger³

et al. 2021

Preprint

View full text Add to dashboard Cite

Half of long-term care (LTC) residents are malnourished increasing hospitalization, mortality, morbidity, with lower quality of life. Current tracking methods are subjective and time consuming. This paper presents the automated food imaging and nutrient intake tracking (AFINI-T) technology designed for LTC. We propose a novel convolutional autoencoder for food classification, trained on an augmented UNIMIB2016 dataset and tested on our simulated LTC food intake dataset (12 meal scenarios; up to 15 classes each; top-1 classification accuracy: 88.9%; mean intake error: -0.4 mL±36.7 mL). Nutrient intake estimation by volume was strongly linearly correlated with nutrient estimates from mass (r 2 0.92 to 0.99) with good agreement between methods (σ = −2.7 to −0.01; zero within each of the limits of agreement). The AFINI-T approach is a deep-learning powered computational nutrient sensing system that may provide a novel means for more accurately and objectively tracking LTC resident food intake to support and prevent malnutrition tracking strategies.

show abstract

Automatic Food Intake Monitoring Based on Chewing Activity: A Survey

Cited by 27 publications

References 90 publications

Smartwatch-Based Eating Detection: Data Selection for Machine Learning from Imbalanced Data with Imperfect Labels

Smartwatch-Based Eating Detection: Data Selection for Machine Learning from Imbalanced Data with Imperfect Labels

Wearable Hybrid Device Capable of Interactive Perception with Pressure Sensing and Visualization

Enhancing Food Intake Tracking in Long-Term Care with Automated Food Imaging and Nutrient Intake Tracking (AFINI-T) Technology

Contact Info

Product

Resources

About