Nabil Alshurafa

Designing a Robust Activity Recognition Framework for Health and Exergaming Using Wearable Sensors

Detecting human activity independent of intensity is essential in many applications, primarily in calculating metabolic equivalent rates and extracting human context awareness. Many classifiers that train on an activity at a subset of intensity levels fail to recognize the same activity at other intensity levels. This demonstrates weakness in the underlying classification method. Training a classifier for an activity at every intensity level is also not practical. In this paper, we tackle a novel intensity-independent activity recognition problem where the class labels exhibit large variability, the data are of high dimensionality, and clustering algorithms are necessary. We propose a new robust stochastic approximation framework for enhanced classification of such data. Experiments are reported using two clustering techniques, K-Means and Gaussian Mixture Models. The stochastic approximation algorithm consistently outperforms other well-known classification schemes which validate the use of our proposed clustered data representation. We verify the motivation of our framework in two applications that benefit from intensity-independent activity recognition. The first application shows how our framework can be used to enhance energy expenditure calculations. The second application is a novel exergaming environment aimed at using games to reward physical activity performed throughout the day, to encourage a healthy lifestyle.

A dense pressure sensitive bedsheet design for unobtrusive sleep posture monitoring

Sleep plays a pivotal role in the quality of life, and sleep posture is related to many medical conditions such as sleep apnea. In this paper, we design a dense pressure-sensitive bedsheet for sleep posture monitoring. In contrast to existing techniques, our bedsheet system offers a completely unobtrusive method using comfortable textile sensors. Based on high-resolution pressure distributions from the bedsheet, we develop a novel framework for pressure image analysis to monitor sleep postures, including a set of geometrical features for sleep posture characterization and three sparse classifiers for posture recognition. We run a pilot study and evaluate the performance of our methods with 14 subjects to analyze 6 common postures. The experimental results show that our proposed method enables reliable sleep posture recognition and offers better overall performance than state-of-the-art methods, achieving up to 83.0% precision and 83.2% recall on average.

Monitoring eating habits using a piezoelectric sensor-based necklace

Maintaining appropriate levels of food intake and developing regularity in eating habits is crucial to weight loss and the preservation of a healthy lifestyle. Moreover, awareness of eating habits is an important step towards portion control and weight loss. In this paper, we introduce a novel food-intake monitoring system based around a wearable wireless-enabled necklace. The proposed necklace includes an embedded piezoelectric sensor, small Arduino-compatible microcontroller, Bluetooth LE transceiver, and Lithium-Polymer battery. Motion in the throat is captured and transmitted to a mobile application for processing and user guidance. Results from data collected from 30 subjects indicate that it is possible to detect solid and liquid foods, with an F-measure of 0.837 and 0.864, respectively, using a naive Bayes classifier. Furthermore, identification of extraneous motions such as head turns and walking are shown to significantly reduce the false positive rate of swallow detection.

A Wearable Nutrition Monitoring System

Maintaining appropriate levels of food intake anddeveloping regularity in eating habits is crucial to weight lossand the preservation of a healthy lifestyle. Moreover, maintainingawareness of ones own eating habits is an important steptowards portion control and ultimately, weight loss. Though manysolutions have been proposed in the area of physical activitymonitoring, few works attempt to monitor an individuals foodintake by means of a noninvasive, wearable platform. In thispaper, we introduce a novel nutrition-intake monitoring systembased around a wearable, mobile, wireless-enabled necklacefeaturing an embedded piezoelectric sensor. We also propose aframework capable of estimating volume of meals, identifyinglong-term trends in eating habits, and providing classificationbetween solid foods and liquids with an F-Measure of 85% and86% respectively. The data is presented to the user in the formof a mobile application.

Determining the Single Best Axis for Exercise Repetition Recognition and Counting on SmartWatches

Due to the exploding costs of chronic diseasesstemming from physical inactivity, wearable sensor systems toenable remote, continuous monitoring of individuals has increasedin popularity. Many research and commercial systems exist inorder to track the activity levels of users from general dailymotion to detailed movements. This work examines this problemfrom the space of smartwatches, using the Samsung GalaxyGear, a commercial device containing an accelerometer and agyroscope, to be used in recognizing physical activity. This workalso shows the sensors and features necessary to enable suchsmartwatches to accurately count, in real-time, the repetitions offree-weight and body-weight exercises. The goal of this work isto try and select only the best single axis for each activity byextracting only the most informative activity-specific features, inorder to minimize computational load and power consumptionin repetition counting. The five activities are incorporated in aworkout routine, and knowing this information, a random forestclassifier is built with average area under the curve (AUC) of: 974, with average accuracy of 93%, in cross validation to identify eachrepetition of a given exercise using all available sensors and AUCof: 950 with accuracy of 89.9% using the single best axis foreach activity alone. Adding a gyroscope with the accelerometerincreased the average AUC from: 968 to: 974, increasing theaccuracy of specific movements as much as 2%. Results show that, while a combination of accelerometer and gyroscope provide thestrongest classification results, often times features extracted froma single, best axis are enough to accurately identify movementsfor a personal training routine, where that axis is often, but notalways, an accelerometer axis.

Recognition of Nutrition Intake Using Time-Frequency Decomposition in a Wearable Necklace Using a Piezoelectric Sensor

Food intake levels, hydration, ingestion rate, and dietary choices are all factors known to impact the risk of obesity. This paper presents a novel wearable system in the form of a necklace, which aggregates data from an embedded piezoelectric sensor capable of detecting skin motion in the lower trachea during ingestion. The skin motion produces an output voltage with varying frequencies over time. As a result, we propose an algorithm based on time-frequency decomposition, spectrogram analysis of piezoelectric sensor signals, to accurately distinguish between food types, such as liquid and solid, hot and cold drinks, and hard and soft foods. The necklace transmits data to a smartphone, which performs the processing of the signals, classifies the food type, and provides visual feedback to the user to assist the user in monitoring their eating habits over time. We compare our spectrogram analysis with other time-frequency features, such as matching pursuit and wavelets. Experimental results demonstrate promise in using time-frequency features, with high accuracy of distinguishing between food categories using spectrogram analysis and extracting key features representative of the unique swallow patterns of various foods.

Robust human intensity-varying activity recognition using Stochastic Approximation in wearable sensors

Detecting human activity independent of intensity is essential in many applications, primarily in calculating metabolic equivalent rates (MET) and extracting human context awareness from on-body inertial sensors. Many classifiers that train on an activity at a subset of intensity levels fail to classify the same activity at other intensity levels. This demonstrates weakness in the underlying activity model. Training a classifier for an activity at every intensity level is also not practical. In this paper we tackle a novel intensity-independent activity recognition application where the class labels exhibit large variability, the data is of high dimensionality, and clustering algorithms are necessary. We propose a new robust Stochastic Approximation framework for enhanced classification of such data. Experiments are reported for each dataset using two clustering techniques, K-Means and Gaussian Mixture Models. The Stochastic Approximation algorithm consistently outperforms other well-known classification schemes which validates the use of our proposed clustered data representation.

Sleep posture analysis using a dense pressure sensitive bedsheet

Sleep posture affects the quality of our sleep and is especially important for such medical conditions as sleep apnea and pressure ulcers. In this paper, we propose a design for a dense pressure-sensitive bedsheet along with an algorithmic framework to recognize and monitor sleeping posture. The bedsheet system uses comfortable textile sensors that produces high-resolution pressure maps. We develop a novel framework for pressure image analysis to monitor sleep postures, including a set of geometrical features for sleep posture characterization and three sparse classifiers for posture recognition. In demonstrating this system, we run 2 pilot studies: one evaluates the performance of our methods with 14 subjects to analyze 6 common postures; the other is a series of overnight studies to verify continuous performance. The experimental results show that our proposed method enables reliable sleep posture recognition and offers better overall performance than traditional methods, achieving up to 83.0% precision and 83.2% recall on average.

Battery optimization in smartphones for remote health monitoring systems to enhance user adherence

Remote health monitoring (RHM) can help save the cost burden of unhealthy lifestyles. Of increased popularity is the use of smartphones to collect data, measure physical activity, and provide coaching and feedback to users. One challenge with this method is to improve adherence to prescribed medical regimens. In this paper we present a new battery optimization method that increases the battery lifetime of smartphones which monitor physical activity. We designed a system, WANDA-CVD, to test our battery optimization method. The focus of this report describes our in-lab pilot study and a study aimed at reducing cardiovascular disease (CVD) in young women, the Womens Heart Health study. Conclusively, our battery optimization technique improved battery lifetime by 300%. This method also increased participant adherence to the remote health monitoring system in the Womens Heart Health study by 53%.

Remote patient monitoring: what impact can data analytics have on cost?

While significant effort has been made on designing Remote Monitoring Systems (RMS), limited research has been conducted on the potential cost savings that these systems offer in terms of reduction in readmission costs, as well as the costs associated with human resources involved in the intervention process. This paper is particularly interested in exploring potential cost savings that an analytics engine can provide in presence of intelligent back-end data processing and machine learning algorithms against conventional RMS that operate based on simple thresholding approaches. Using physiological data collected from 486 heart failure patients through a clinical study in collaboration with the UCLA School of Medicine, we conduct a retrospective data analysis to estimate prediction accuracy as well as associated costs of the two remote monitoring approaches. Our results show that analytics-based RMS can reduce false negative rates by 61.4% while maintaining a false positive performance close to that of conventional RMS. Furthermore, the proposed analytics engine achieves 61.5% reduction in the overall readmission costs.