Konrad Lorincz

Sensor networks for emergency response: challenges and opportunities

Sensor networks, a new class of devices has the potential to revolutionize the capture, processing, and communication of critical data for use by first responders. CodeBlue integrates sensor nodes and other wireless devices into a disaster response setting and provides facilities for ad hoc network formation, resource naming and discovery, security, and in-network aggregation of sensor-produced data. We designed CodeBlue for rapidly changing, critical care environments. To test it, we developed two wireless vital sign monitors and a PDA-based triage application for first responders. Additionally, we developed MoteTrack, a robust radio frequency (RF)-based localization system, which lets rescuers determine their location within a building and track patients. Although much of our work on CodeBlue is preliminary, our initial experience with medical care sensor networks raised many exciting opportunities and challenges.

Sensor networks for medical care

Sensor networks have the potential to greatly impact many aspects of medical care. By outfitting patients with wireless, wearable vital sign sensors, collecting detailed real-time data on physiological status can be greatly simplified. However, there is a significant gap between existing sensor network systems and the needs of medical care. In particular, medical sensor networks must support multicast routing topologies, node mobility, a wide range of data rates and high degrees of reliability, and security. This paper describes our experiences with developing a combined hardware and software platform for medical sensor networks, called CodeBlue. CodeBlue provides protocols for device discovery and publish/subscribe multihop routing, as well as a simple query interface that is tailored for medical monitoring. We have developed several medical sensors based on the popular MicaZ and Telos mote designs, including a pulse oximeter, EKG and motion-activity sensor. We also describe a new, miniaturized sensor mote designed for medical use. We present initial results for the CodeBlue prototype demonstrating the integration of our medical sensors with the publish/subscribe routing substrate. We have experimentally validated the prototype on our 30-node sensor network testbed, demonstrating its scalability and robustness as the number of simultaneous queries, data rates, and transmitting sensors are varied. We also study the effect of node mobility, fairness across multiple simultaneous paths, and patterns of packet loss, confirming the system’s ability to maintain stable routes despite variations in node location and

Mercury: a wearable sensor network platform for high-fidelity motion analysis

This paper describes Mercury, a wearable, wireless sensor platform for motion analysis of patients being treated for neuromotor disorders, such as Parkinsons Disease, epilepsy, and stroke. In contrast to previous systems intended for short-term use in a laboratory, Mercury is designed to support long-term, longitudinal data collection on patients in hospital and home settings. Patients wear up to 8 wireless nodes equipped with sensors for monitoring movement and physiological conditions. Individual nodes compute high-level features from the raw signals, and a base station performs data collection and tunes sensor node parameters based on energy availability, radio link quality, and application specific policies. Mercury is designed to overcome the core challenges of long battery lifetime and high data fidelity for long-term studies where patients wear sensors continuously 12 to 18 hours a day. This requires tuning sensor operation and data transfers based on energy consumption of each node and processing data under severe computational constraints. Mercury provides a high-level programming interface that allows a clinical researcher to rapidly build up different policies for driving data collection and tuning sensor lifetime. We present the Mercury architecture and a detailed evaluation of two applications of the system for monitoring patients with Parkinsons Disease and epilepsy.

The Advanced Health and Disaster Aid Network: A Light-Weight Wireless Medical System for Triage

Advances in semiconductor technology have resulted in the creation of miniature medical embedded systems that can wirelessly monitor the vital signs of patients. These lightweight medical systems can aid providers in large disasters who become overwhelmed with the large number of patients, limited resources, and insufficient information. In a mass casualty incident, small embedded medical systems facilitate patient care, resource allocation, and real-time communication in the advanced health and disaster aid network (AID-N). We present the design of electronic triage tags on lightweight, embedded systems with limited memory and computational power. These electronic triage tags use noninvasive, biomedical sensors (pulse oximeter, electrocardiogram, and blood pressure cuff) to continuously monitor the vital signs of a patient and deliver pertinent information to first responders. This electronic triage system facilitates the seamless collection and dissemination of data from the incident site to key members of the distributed emergency response community. The real-time collection of data through a mesh network in a mass casualty drill was shown to approximately triple the number of times patients that were triaged compared with the traditional paper triage system.

Wireless Medical Sensor Networks in Emergency Response: Implementation and Pilot Results

This project demonstrates the feasibility of using cost- effective, flexible, and scalable sensor networks to address critical bottlenecks of the emergency response process. For years, emergency medical service providers conducted patient care by manually measuring vital signs, documenting assessments on paper, and communicating over handheld radios. When disasters occurred, the large numbers of casualties quickly and easily overwhelmed the responders. Collaboration with EMS and hospitals in the Baltimore Washington Metropolitan region prompted us to develop miTag (medical information tag), a cost- effective wireless sensor platform that automatically track patients throughout each step of the disaster response process, from disaster scenes, to ambulances, to hospitals. The miTag is a highly extensible platform that supports a variety of sensor add-ons - GPS, pulse oximetry, blood pressure, temperature, ECG - and relays data over a self-organizing wireless mesh network Scalability is the distinguishing characteristic of miTag: its wireless network scales across a wide range of network densities, from sparse hospital network deployments to very densely populated mass casualty sites. The miTag system is out-of-the-box operational and includes the following key technologies: 1) cost-effective sensor hardware, 2) self-organizing wireless network and 3) scalable server software that analyzes sensor data and delivers real-time updates to handheld devices and web portals. The system has evolved through multiple iterations of development and pilot deployments to become an effective patient monitoring solution. A pilot conducted with the Department of Homeland Security indicates miTags can increase the patient care capacity of responders in the field A pilot at Washington Hospital showed miTags are capable of reliably transmitting data inside radio-interference-rich critical care settings.

MoteTrack: a robust, decentralized approach to RF-Based location tracking

Wireless transmitters deployed throughout an indoor environment offer the opportunity for accurate location tracking of mobile users. Using radio signal information alone, it is possible to determine the location of a roaming node at close to meter-level accuracy. We are particularly concerned with applications in which the robustness of the locationtracking infrastructure is at stake. For example, firefighters and rescuers entering a building can use a heads-up display to track their location and monitor safe exit routes. Likewise, an incident commander could track the location of multiple rescuers in the building from the command post. In this paper, we present a robust, decentralized approach to RFbased location tracking. Our system, called MoteTrack, is based on low-power radio transceivers coupled with a modest amount of computation and storage capabilities. MoteTrack does not rely upon any back-end server or network infrastructure: the location of each mobile node is computed using a received radio signal strength signature from numerous beacon nodes to a database of signatures that is replicated across the beacon nodes themselves. This design allows the system to function despite significant failures of the radio beacon infrastructure. In our deployment of MoteTrack, consisting of 20 beacon nodes distributed across our Computer Science building, we achieve a 50 percentile and 80 percentile location-tracking accuracy of 2 meters and 3 meters respectively. In addition, MoteTrack can tolerate the failure of up to 60% of the beacon nodes without severely degrading accuracy, making the system suitable for deployment in highly volatile conditions. We present a detailed analysis of MoteTrack’s performance under a wide range of conditions, including variance in the number of obstructions, beacon node failure, radio signature perturbations, receiver sensitivity, and beacon node density.

Resource aware programming in the Pixie OS

This paper presents Pixie, a new sensor node operating system designed to support the needs of data-intensive applications. These applications, which include high-resolution monitoring of acoustic, seismic, acceleration, and other signals, involve high data rates and extensive in-network processing. Given the fundamentally resource-limited nature of sensor networks, a pressing concern for such applications is their ability to receive feedback on, and adapt their behavior to, fluctuations in both resource availability and load. The Pixie OS is based on a dataflow programming model based on the concept of resource tickets, a core abstraction for representing resource availability and reservations. By giving the system visibility and fine-grained control over resource management, a broad range of policies can be implemented. To shield application programmers from the burden of managing these details, Pixie provides a suite of resource brokers, which mediate between low-level physical resources and higher-level application demands. Pixie is implemented in NesC and supports limited backwards compatibility with TinyOS. We describe Pixie in the context of two applications: limb motion analysis for patients undergoing treatment for motion disorders, and acoustic target detection using a network of microphones. We present a range of experiments demonstrating Pixies ability to accurately account for resource availability at runtime and enable a range of both generic and application-specific adaptations.

Analysis of Feature Space for Monitoring Persons with Parkinson's Disease With Application to a Wireless Wearable Sensor System

We present work to develop a wireless wearable sensor system for monitoring patients with Parkinsons disease (PD) in their homes. For monitoring outside the laboratory, a wearable system must not only record data, but also efficiently process data on-board. This manuscript details the analysis of data collected using tethered wearable sensors. Optimal window length for feature extraction and feature ranking were calculated, based on their ability to capture motor fluctuations in persons with PD. Results from this study will be employed to develop a software platform for the wireless system, to efficiently process on-board data.

Wearable wireless sensor network to assess clinical status in patients with neurological disorders

The goal of this project is to develop wireless sensors and analysis methods to monitor patients with various motor dysfunctions. We are currently targeting two specific applications: facilitating medication titration in patients with Parkinsons disease and assessing motor recovery in stroke survivors undergoing rehabilitation. In our vision, the treatment and rehabilitation hospital of the future will allow clinicians to continuously monitor motor activity in patients via miniature sensor technology in order to better design interventions on an individual basis. Two key points toward developing the tools necessary to achieve continuous monitoring of motor function are (1) development of a robust and deployable wearable wireless network of sensors and (2) the development of analysis techniques to derive clinically relevant information from miniature sensor data.

Sensor networks for high-resolution monitoring of volcanic activity

We developed and deployed a wireless sensor network for monitoring seismoacoustic activity at Volcán Reventador, Ecuador. Wireless sensor networks are a new technology and our group is among the first to apply them to monitoring volcanoes. The small size, low power, and wireless communication capabilities can greatly simplify deployments of large sensor arrays and are very attractive for this application domain. This project is a follow-on to our previous infrasonic sensor network deployed at Volcán Tungurahua, also in Ecuador, in July 2004 [1].