Ramin Fallahzadeh

An Energy-Efficient Computational Model for Uncertainty Management in Dynamically Changing Networked Wearables

The utility of wearables is currently limited to lab experiments and controlled environments mainly because computational algorithms embedded in wearables fail to produce accurate measurements in uncontrolled, dynamically changing, and potentially harsh environments. With the exponentially growing adoption of these systems in human-centered Internet-of-Things (IoT) applications, development of resource-efficient solutions to enhance the accuracy of this systems remains a considerable research challenge. In this paper, we introduce an energy-efficient framework for uncertainty management of networked wearables. The core components of our framework are anomaly screening units for detecting anomalies that require handling, thus resulting in one order of magnitude less energy consumption compared to the conventional frameworks. Furthermore, our screening approach achieves 98.3% accuracy in detecting anomalies based on real data collected with wearable motion sensors.

A Reliable and Reconfigurable Signal Processing Framework for Estimation of Metabolic Equivalent of Task in Wearable Sensors

Wearable motion sensors are widely used to estimate metabolic equivalent of task (MET) values associated with physical activities. However, one major obstacle in widespread adoption of current wearables is that any changes in configuration of the network requires new data collection and re-training of the underlying signal processing algorithms. For any wearable-based MET estimation framework to be considered a viable platform, it needs to be reconfigurable, reliable, and power-efficient. In this paper, we aim to address the issues of sensor misplacement, power efficiency, and new sensor addition and propose a reliable and reconfigurable MET estimation framework. We introduce a power-aware sensor localization approach that allows users to wear the sensors on different body locations without need for adhering to a specific installation protocol. Furthermore, we propose a novel transductive transfer learning approach, which gives end-users the ability to add new sensors to the network without need for collecting new training data. This is accomplished by transferring the knowledge of already trained sensors to the untrained sensors in real-time. Our experiments demonstrate that our sensor localization algorithm achieves an accuracy of 90.8% in detecting location of the wearable sensors. The integrated model of sensor localization and MET calculation achieves an R2 of 0.8 in estimating MET values using a regression-based model. Furthermore, our transfer learning algorithm improves the R2 value of MET estimation up to 60%.

Smart-Cuff: A wearable bio-sensing platform with activity-sensitive information quality assessment for monitoring ankle edema

Leg swelling produced by retention of fluid in leg tissues is known as peripheral edema, which is regarded as a symptom for various systematic diseases such as heart or kidney failure. In current clinical practice, edema is manually assessed by clinical experts. Such an assessment can often be inaccurate and unreliable especially if it is made by different operators at different times. Despite the importance of monitoring edema for the purpose of evaluating the course of disease or the effect of treatment, quantifying peripheral edema in a continuous and accurate fashion has remained a challenge. In this paper, we propose a wearable real-time platform (namely, Smart-Cuff), which integrates advanced technologies in sensing, computation, and signal processing and machine learning for continuous and real-time edema monitoring in remote and in-home settings. Given that peripheral edema is highly dependent on various contextual attributes such as body posture, we present an activity-sensitive approach to discard erroneous or contextually invalid sensor data in order to meet the requirements of both energy efficiency and quality of information. Examination of our hardware prototype demonstrates the effectiveness of the proposed force-sensitive resistor-based edema sensor (with an R2 of 0.97 for our regression model) as well as the activity monitoring mechanism (over 99% accuracy) that provide the means to perform reliable data sanity check on ankle circumference measurements in a continuous manner.

Toward personalized and context-aware prompting for smartphone-based intervention

Intervention strategies can help individuals with cognitive impairment to increase adherence to instructions, independence, and activity engagement and reduce errors on everyday instrumental activities of daily living (IADLs) and caregiver burden. However, to be effective, intervention prompts should be given at a time that does not interrupt other important user activities and is more convenient. In this paper, we propose an intelligent personalized intervention system for smartphones. In our approach, we use context and activity awareness to time prompts when they will most likely be viewed and used. Our result based on real data collected using smartphone motion sensors demonstrate that the proposed approach can detect the time-frame of a user response with an average accuracy of 65% and reduce the inefficiency by 39%, on average, compared to different static time interventions which shows the possibilities and advantages of the proposed system to increase user satisfaction and response rate.

A Hardware-Assisted Energy-Efficient Processing Model for Activity Recognition Using Wearables

Wearables are being widely utilized in health and wellness applications, primarily due to the recent advances in sensor and wireless communication, which enhance the promise of wearable systems in providing continuous and real-time monitoring and interventions. Wearables are generally composed of hardware/software components for collection, processing, and communication of physiological data. Practical implementation of wearable monitoring in real-life applications is currently limited due to notable obstacles. The wearability and form factor are dominated by the amount of energy needed for sensing, processing, and communication. In this article, we propose an ultra-low-power granular decision-making architecture, also called screening classifier, which can be viewed as a tiered wake-up circuitry, consuming three orders of magnitude-less power than the state-of-the-art low-power microcontrollers. This processing model operates based on computationally simple template matching modules, based on coarse- to fine-grained analysis of the signals with on-demand and gradually increasing the processing power consumption. Initial template matching rejects signals that are clearly not of interest from the signal processing chain, keeping the rest of processing blocks idle. If the signal is likely of interest, the sensitivity and the power of the template matching modules are gradually increased, and ultimately, the main processing unit is activated. We pose optimization techniques to efficiently split a full template into smaller bins, called mini-templates, and activate only a subset of bins during each classification decision. Our experimental results on real data show that this signal screening model reduces power consumption of the processing architecture by a factor of 70% while the sensitivity of detection remains at least 80%.

A machine learning approach for medication adherence monitoring using body-worn sensors

One of the most important challenges in chronic disease self-management is medication non-adherence, which has irrevocable outcomes. Although many technologies have been developed for medication adherence monitoring, the reliability and cost-effectiveness of these approaches are not well understood to date. This paper presents a medication adherence monitoring system by user-activity tracking based on wrist-band wearable sensors. We develop machine learning algorithms that track wrist motions in real-time and identify medication intake activities. We propose a novel data analysis pipeline to reliably detect medication adherence by examining single-wrist motions. Our system achieves an accuracy of 78.3% in adherence detection without need for medication pillboxes and with only one sensor worn on either of the wrists. The accuracy of our algorithm is only 7.9% lower than a system with two sensors that track motions of both wrists.

Toward robust and platform-agnostic gait analysis

Biometric gait analysis using wearable sensors offers an objective and quantitative method for gait parameter extraction. However, current techniques are constrained to specific platform parameters, and hence significantly lack generality, scalability and sustainability. In this paper, we propose a platform-independent and self-adaptive approach for gait cycle detection and cadence estimation. Our algorithm utilizes physical kinematic properties and cyclic patterns of foot acceleration signals to automatically adjust internal parameters of the algorithm. As a result, the proposed approach is robust to noise and changes in sensor platform parameters such as sampling rate and sensor resolution. For the evaluation purpose, we use acceleration signals collected from 16 subjects in a clinical setting to examine the accuracy and robustness of the proposed algorithm. The results show that our approach achieves a precision above 98% and a recall above 95% in stride detection, and an average accuracy of 98% in cadence estimation under various uncertainty conditions such as noisy signals and changes in sampling frequency and sensor resolution.

A novel test strategy and fault-tolerant routing algorithm for NoC routers

In this paper, we present a novel routing algorithm in order to avoid deadlock and packet dropping. In our proposed algorithm the network-on-chip (NoC) is capable of tolerating faults in presence of control faults in combinational parts of routers. In addition, by modifying the functionality of the router, the router is enabled to test its own, as well as the preceding routers functionality based on the routing algorithm, destination address and previous routers situation. Each router recognizes the faulty neighbor and announces it to successive routers. In this scheme no extra packets will be generated. We analyze the effects of our method on latency, power consumption and drop rate. Our experimental results illustrate that, fault coverage for routers can reach up to 100% with yet low power consumption and significant improvement in latency compared to the baseline approach.

SmartSock: A wearable platform for context-aware assessment of ankle edema

Ankle edema an important symptom for monitoring patients with chronic systematic diseases. It is an important indicator of onset or exacerbation of a variety of diseases that disturb cardiovascular, renal, or hepatic system such as heart, liver, and kidney failure, diabetes, etc. The current approaches toward edema assessment are conducted during clinical visits. In-clinic assessments, in addition to being burdensome and expensive, are sometimes not reliable and neglect important contextual factors such as patients physical activity level and body posture. A novel wearable sensor, namely SmartSock, equipped with accelerometer and flexible stretch sensor embedded in clothing is presented. SmartSock is powered by advanced machine learning, signal processing, and correlation techniques to provide real-time, reliable, and context-rich information in remote settings. Our experiments on human subjects indicate high confidence in activity and posture recognition (with an accuracy of > 96%) as well as reliable edema quantification with intra-class correlation and Pearson correlation of 0.97.Ankle edema an important symptom for monitoring patients with chronic systematic diseases. It is an important indicator of onset or exacerbation of a variety of diseases that disturb cardiovascular, renal, or hepatic system such as heart, liver, and kidney failure, diabetes, etc. The current approaches toward edema assessment are conducted during clinical visits. In-clinic assessments, in addition to being burdensome and expensive, are sometimes not reliable and neglect important contextual factors such as patients physical activity level and body posture. A novel wearable sensor, namely SmartSock, equipped with accelerometer and flexible stretch sensor embedded in clothing is presented. SmartSock is powered by advanced machine learning, signal processing, and correlation techniques to provide real-time, reliable, and context-rich information in remote settings. Our experiments on human subjects indicate high confidence in activity and posture recognition (with an accuracy of > 96%) as well as reliable edema quantification with intra-class correlation and Pearson correlation of 0.97.

Glaucoma-Specific Gait Pattern Assessment Using Body-Worn Sensors

Many studies have reported that glaucoma patients experience mobility issues, such as walking slowly and bumping into obstacles frequently. However, little is known to date about how a persons gait is impacted due to glaucoma. This paper presents design and development of a gait analysis approach using a shoe-integrated sensing system and accompanying machine learning techniques to quantitatively examine gait patterns in glaucoma patients. The customized sensor platform is utilized in a clinical trial conducted with nine glaucoma patients and ten age-matched healthy participants. The signal processing and machine learning algorithms automatically detect effective gait cycles and extract both steady-state and spatio-temporal gait features from the signal segments. We perform machine learning algorithms to distinguish glaucoma patients from healthy controls, and identify several prominent features with high discriminability between the two groups. The results demonstrate that classification algorithms can be used to identify the gait patterns of glaucoma patients with an accuracy higher than 94% in a 10-m-walk test. It is also demonstrated that gait features such as evenness of the sway speed along medio-lateral direction between the two feet are significantly different (p-value <; 0.001) between older adults with and without glaucoma. These results suggest that emerging solutions, such as wearable sensing technologies, can be used for continuous and real-time assessment of gait and mobility problems in individuals with low vision, and may open new avenues for using changes in gait patterns for preventing life threatening situations such as falls.