CSAIL Publications and Digital Archive header
bullet Technical Reports bullet Work Products bullet Research Abstracts bullet Historical Collections bullet

link to publications.csail.mit.edu link to www.csail.mit.edu horizontal line


Research Abstracts - 2006
horizontal line

horizontal line

vertical line
vertical line

SMART: Scalable Medical Alert and Response Technologies

Dorothy Curtis, Eugene Shih, Asfandyar Qureshi, Esteban Pino, Lucila Ohno-Machado, Robert Greenes & John Guttag


The Scalable Medical Alert and Response Technology Project (SMART) [1] project examines deploying advanced technology in the waiting area of an Emergency Department to wirelessly monitor vital signs and locations of otherwise unattended patients. This will give waiting patients some security that their condition is being monitored even though a caregiver is unavailable. The Emergency Department caregivers will be alerted to problems occurring in the waiting room in real-time, while alerts and patient priorities can be dynamically adjusted depending on the circumstances. We hope to gain some insights from this Emergency Department deployment with goal of being able to quickly deploy this system in disaster situations where patients drastically outnumber available caregivers or in mobile hospitals at disaster sites.


In this project we plan to monitor patients in the waiting room who have an initial complaint of chest pain. We will give each patient a handheld computer and two sensors. The sensors, which will be monitoring the electrical rhythms of the heart (ECG) and the oxygenation levels (SpO2), will send their data to the patient's handheld. The handheld will forward the data to a central server. The central server will monitor the signals and alert appropriate caregivers when there is a problem. The system also has an indoor positioning system manufactured by Sonitor [2], for locating patients, caregivers, and equipment.


During the past year we refined our prototype of this system: we now have eight working patient PDAs with sensor boxes and two caregiver PDAs and have tested the system using healthy volunteers.

Our prototype has four components: a wearable patient monitor, a central computer, a caregiver interface and a tracking system. The patient monitor is based on an HP iPAQ model 5500. This PDA interfaces to a sensor box. The sensor box provides physiological signals from a 1-lead ECG sensor and an SpO2 sensor to the PDA.  The PDA forwards this data wirelessly to the SMART Central computer.

Figure 1: SMART System Components
SMART Central includes a Streaming Data Manager (SDM), a Decision Support Module (DSM), and a Logistics Support Module (LSM).

The SMART Central computer analyzes the patient data for alarm conditions. Alarm conditions include high heart rate, low heart rate and low SpO2. When an alarm condition is detected, it is dispatched to an available caregiver.  There are also “technical” alarms for when a sensor is disconnected.

This prototype system also gives each caregiver a PDA. These PDAs allow the caregivers to see the roster of patients and to click through to see a patient's vital signs in real-time. The PDAs built-in vibration feature is used to alert the caregiver PDA has received an alarm.

To achieve our goals for tracking patients and caregivers, we are using the Sonitor location system. This system consists of tags and detectors.  Each patient, caregiver, and piece of equipment has a tag, shaped like a large pen.  Each tag emits ultrasound messages with its ID.  Detectors are placed on the walls.  The detectors receive the ultrasound messages from the tags and forward them to SMART Central.  SMART Central places each patient and caregiver in the zone surrounding the nearest detector.

We presented a theatre-style demo [3] at the AMIA 2005 conference and a paper on Real-Time ECG Signal Analysis [4].

We are currently evaluating the system on healthy volunteers in the waiting area of the emergency department at the Brigham and Women’s Hospital.


In the near future we will be conducting an evaluation of our system with real patients.  We have ongoing research projects in the following areas: real-time analysis of mobile ECG signals, managing energy consumption on tiny wireless sensors, transmitting vitals signs data and video from an ambulance via a cellular network, asset tracking, and alarm management.

Research Support

This research is supported by the National Library of Medicine.


[1] Lucila Ohno-Machado and Dorothy Curtis. Smart: Scalable Medical Alert and Response Technology.

[2] Sonitor Technologies: Indoor Positioning System. http://sonitor.com

[3] Demonstration of SMART (Scalable Medical Alert Response Technology). Jason Waterman, Dorothy Curtis, Michel Goraczko, Eugene Shih, Pankaj Sarin MD, Esteban Pino, Lucila Ohno-Machado MD, PhD, Robert Greenes MD, PhD, John Guttag PhD, Thomas Stair MD. Presented at AMIA, Washington, D.C., October 2005.

 [4] Real-Time ECG Algorithms for Ambulatory Patient Monitoring. Esteban Pino, Lucila Ohno-Machado MD, PhD, Eduardo Wiechmann PhD, and Dorothy Curtis. Presented at AMIA, Washington, D.C., October 2005.

vertical line
vertical line
horizontal line

MIT logo Computer Science and Artificial Intelligence Laboratory (CSAIL)
The Stata Center, Building 32 - 32 Vassar Street - Cambridge, MA 02139 - USA
tel:+1-617-253-0073 - publications@csail.mit.edu