NEWERTRACK: ML-Based Accurate Tracking of In-Mouth Nutrient Sensors Position Using Spectrum-Wide Information

Zargari, Amir Hosein Afandizadeh; Dautta, Manik; Ashrafiamiri, Marzieh; Seo, Minjun; Tseng, Peter; Kurdahi, Fadi

doi:10.1109/tcad.2020.3013074

Cited by 6 publications

(3 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning techniques and neural networks have been widely used in health monitoring domains. Zargari et al [ 22 ] used a combination of convolutional neural networks and recurrent neural networks to accurately track the position of in-mouth nutrient sensors. Mehrabadi et al [ 23 ] used convolutional neural networks to detect COVID-19 in patients with acute respiratory distress syndrome.…”

Section: Introductionmentioning

confidence: 99%

Pain Assessment Tool With Electrodermal Activity for Postoperative Patients: Method Validation Study

Aqajari¹,

Cao²,

Naeini³

et al. 2021

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

Background Accurate, objective pain assessment is required in the health care domain and clinical settings for appropriate pain management. Automated, objective pain detection from physiological data in patients provides valuable information to hospital staff and caregivers to better manage pain, particularly for patients who are unable to self-report. Galvanic skin response (GSR) is one of the physiologic signals that refers to the changes in sweat gland activity, which can identify features of emotional states and anxiety induced by varying pain levels. This study used different statistical features extracted from GSR data collected from postoperative patients to detect their pain intensity. To the best of our knowledge, this is the first work building pain models using postoperative adult patients instead of healthy subjects. Objective The goal of this study was to present an automatic pain assessment tool using GSR signals to predict different pain intensities in noncommunicative, postoperative patients. Methods The study was designed to collect biomedical data from postoperative patients reporting moderate to high pain levels. We recruited 25 participants aged 23-89 years. First, a transcutaneous electrical nerve stimulation (TENS) unit was employed to obtain patients' baseline data. In the second part, the Empatica E4 wristband was worn by patients while they were performing low-intensity activities. Patient self-report based on the numeric rating scale (NRS) was used to record pain intensities that were correlated with objectively measured data. The labels were down-sampled from 11 pain levels to 5 different pain intensities, including the baseline. We used 2 different machine learning algorithms to construct the models. The mean decrease impurity method was used to find the top important features for pain prediction and improve the accuracy. We compared our results with a previously published research study to estimate the true performance of our models. Results Four different binary classification models were constructed using each machine learning algorithm to classify the baseline and other pain intensities (Baseline [BL] vs Pain Level [PL] 1, BL vs PL2, BL vs PL3, and BL vs PL4). Our models achieved higher accuracy for the first 3 pain models than the BioVid paper approach despite the challenges in analyzing real patient data. For BL vs PL1, BL vs PL2, and BL vs PL4, the highest prediction accuracies were achieved when using a random forest classifier (86.0, 70.0, and 61.5, respectively). For BL vs PL3, we achieved an accuracy of 72.1 using a k-nearest-neighbor classifier. Conclusions We are the first to propose and validate a pain assessment tool to predict different pain levels in real postoperative adult patients using GSR signals. We also exploited feature selection algorithms to find the top important features related to different pain intensities. In...

show abstract

Section: Introductionmentioning

confidence: 99%

Pain Assessment Tool With Electrodermal Activity for Postoperative Patients: Method Validation Study

Aqajari¹,

Cao²,

Naeini³

et al. 2021

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

show abstract

“…Individual sensors are built with specialized materials (both within and around the sensing architecture) and thus rendered selectively sensitive to metrics such as glucose, sugars, salts, fats, pressure, temperature, and more. [25,30,32,35,44] Importantly, these structures are readily tuned to respond/resonate at different wavelengths, and thus occupy individual frequency bands during spectral readout. This occurs by simply varying the thickness of the interlayer.…”

Section: Resultsmentioning

confidence: 99%

Programmable Multiwavelength Radio Frequency Spectrometry of Chemophysical Environments through an Adaptable Network of Flexible and Environmentally Responsive, Passive Wireless Elements

Dautta

Hajiaghajani

et al. 2022

Small Science

Self Cite

View full text Add to dashboard Cite

Readout of multiparametric environmental signals typically uses discrete sensing formats that individually require unique signal conditioning circuitry and/or processing pathways. Here, adaptable sensor networks composed exclusively of passive material architectures that enable spectrometric comonitoring of chemical or physical environmental signals are proposed. Herein, a single radio frequency (RF) reader wirelessly interacts first with an intermediate wireless relay coil—this is tunable in length and can be designed to conform around surfaces. This relay (that is fused on textiles or surfaces) is then wirelessly coupled to arrays of passive RF sensors with individually programmable flexibility/reactivity to environmental signals. Multiple chemical and physical signals can then be monitored within the single spectral readout of a wearable reader. This technique can probe over tunable length scales, and is robust to mechanical disturbances that limit present techniques. As a proof of concept, this approach is used to comonitor chemophysical metrics such as nutrients, temperature, pressure, pH, and more on the skin or in utensils with a single readout. This technique may form a cornerstone of zero‐microelectronic sensor networks.

show abstract

“…This sensor is finally built into a broadside-coupled split ring resonator format [35][36][37][38] and interrogated wirelessly as shown in Figure 1c. The fundamental response of this wireless sensor can be modeled using a simple parallel resistance-inductancecapacitance (RLC) equivalent circuit.…”

Section: Silk Fibroin Biopolymer-interlayer Sensorsmentioning

confidence: 99%

Multiscale, Nano‐ to Mesostructural Engineering of Silk Biopolymer‐Interlayer Biosensors for Continuous Comonitoring of Nutrients in Food

Dautta

Dia

Hajiaghajani

et al. 2021

Adv Materials Technologies

Self Cite

View full text Add to dashboard Cite

growing tissue, and maintaining cellular function. [5][6][7] Despite its importance in health and human well-being, there are currently no tools that can monitor the nutrition content of foods. Current approaches to measure the nutrient content of food largely occur by bomb calorimetry [8,9] as well as fractionation with subsequent mass spectroscopy. [10][11][12] While these methods are accurate and effective, these require significant sample preparation and expensive/large machinery and are thus not suitable for measuring nutrient content direct from foods in on-the-fly settings so critical to many emerging applications.The major nutrients of food include carbohydrates (separated into complex carbohydrates and simple sugars), salts, oils or fats, protein, and finally water itself. These compose the majority of the weight of food, with carbohydrates composing 5-10%, fats or oils at 5-30%, salts at up to 1%, and proteins at up to 10% weight (with water composing the remaining mass and micronutrients at small concentrations). [13,14] Complex carbohydrates and proteins are difficult (if not impossible) to quantify direct from foods because these are not broken down significantly before intake. However, core nutrients such as simple sugars, salts, oils/fats, and water (that exist in molecules close to their metabolized forms) are key potential targets for nutrient sensing platforms. Indeed, the underconsumption/overconsumption of these critical nutrients are linked to a variety of markers of human health including mental state, [15,16] disease onset/maintenance, [17,18] and body development. [19,20] We recently proposed and demonstrated a partially selective biosensor composed of a silk fibroin biopolymer-interlayer interceding two electrodes (structured into a wireless format) for nutrient detection. [21] The silk biopolymer forms a dense, water-absorbent membrane that inherently performs sample preparation on the fluidic environment, while absorbing and/ or swelling in response to salts or simple carbohydrates. This device was attached to a human tooth and used to measure the sugar, salt, and alcohol content from a variety of oral liquids. This sensor had a variety of limitations including relatively slow response times, lack of programmability in sensor Nutrition measurement has broad applications in science, ranging from dietary assessment, to food monitoring, personalized health, and more. Despite its importance, there are currently no tools that offer continuous cotracking of nutrients direct from food. In this study, the multiscale engineering of silk biopolymer-interlayer sensors is reported for comonitoring of nutrients. By manipulating various nano-to mesostructural properties of such biosensors, sensors are obtained with programmable sensitivity and selectivity to salts, sugars, and oils/fats. Notably, this approach requires no specialized nanomaterials or delicate biomolecules. Programmable biosensors are further formatted for wireless readout and characteristics of these passive, wireless nutrient monito...

show abstract

NEWERTRACK: ML-Based Accurate Tracking of In-Mouth Nutrient Sensors Position Using Spectrum-Wide Information

Cited by 6 publications

References 27 publications

Pain Assessment Tool With Electrodermal Activity for Postoperative Patients: Method Validation Study

Pain Assessment Tool With Electrodermal Activity for Postoperative Patients: Method Validation Study

Programmable Multiwavelength Radio Frequency Spectrometry of Chemophysical Environments through an Adaptable Network of Flexible and Environmentally Responsive, Passive Wireless Elements

Multiscale, Nano‐ to Mesostructural Engineering of Silk Biopolymer‐Interlayer Biosensors for Continuous Comonitoring of Nutrients in Food

Contact Info

Product

Resources

About