Method and Apparatus for Communication with Bystanders in the Event of a Catastrophic Personal Emergency

[0001] This application claims the benefit of U.S. provisional application Ser. No. 60/693,645 filed Jun. 24, 2005.

[0002] This invention relates to emergency medical alert devices and, in particular, to a system including a body worn audio communication device which can provide bystanders with verbal instructions in the event of a catastrophic personal health emergency.

[0003] Medical alert devices can save a person's life when the person experiences a catastrophic medical condition in a public area. These devices are of particular use in situations that leave a patient unconscious, such as the onset of sudden cardiac arrest. These devices can respond by communicating with a system that provides it with an indication of personal health, which can be as basic as “ok” or “not ok”. This indication may be manually activated, or may include automated monitoring of one or more patient attributes coupled with automated determination of the patient's state of health.

[0004] Personal emergency response devices are manually activated systems that summon aid, which can include voice or other locally generated audio instructions that would help bystanders provide emergency assistance. One such device is described in U.S. Pat. No. 6,292,687 and includes a heart dysfunction reader and sensor worn by a patient on the chest or wrist. A vital sign is monitored and the sensor determines whether a sign indicates a heart dysfunction. A signal is sent to a loop process which in turn sends a signal to a personal alarm worn by the patient. The personal alarm can broadcast a synthetic or recorded voice alerting bystanders to the medical condition.

[0005] Another device is described in US Pat. Appl. 2005/0065445 which describes an implantable sensor that, upon detection of abnormal heart activity, transmits a radio frequency signal to an external receiver carried by the patient. The external receiver has multiple communication capabilities including an enunciator which may be heard by bystanders.

[0006] Another device is an attitude-activated warning device described in U.S. Pat. No. 3,634,885. The device is worn by the patient and contains a mercury switch when closes if the patient collapses to a prone position. The switch closure activates an endless tape in the device which gives instructions to bystanders by means of a loudspeaker. A similar device is described in U.S. Pat. No. 6,570,503. See also US Pat. Appl. 2005/0030190.

[0007] There remains a need for a body worn audio communication device which monitors a patient vital sign and issues verbal instructions to bystanders in the event of an incapacitating medical emergency. Such a device should be compact, unobtrusive, and exhibit low power consumption.

[0008] In accordance with the principles of the present invention, an audio communication device is described which improves the probability of a patient's survival by providing directions for bystanders in verbal form from a physiological sensor and processor worn on the patient's body. The simple sounding of an alarm tone might cause a bystander to pay attention to an unconscious person, but when the alarm is accompanied by a voice prompt that for example, states “This is a medical emergency. Please call for help immediately”, the bystander will know that the alarm was not a cell phone ringing, but was in fact a call for assistance. Upon hearing the voice instruction, the bystander will be able to recognize the nature of the alarm and is given the next steps to follow, thus reducing overall reaction time and improving patient outcome.

[0017] There are numerous ways to implement an on-body emergency instruction system in accordance with the principles of the present invention. In one example described below the system comprises a device that integrates a physiological signal sensor, processor, and loudspeaker that attaches to the skin and that is unobtrusive such that a wearer is not overly burdened by its presence. Conventional audio transducers for example, tend to be relatively large and bulky if they are to be loud enough to be heard by bystanders. Headphone drivers, for example, are small but may not produce sufficient audio volume to be easily heard at a short distance. Speaker systems consisting of a motor and a moving diaphragm tend to be large, bulky, and require considerable power. These characteristics all conflict with the objective of being unobtrusive.

[0018] The following examples of the present invention provide satisfactory fidelity while emphasizing small size and power efficiency. These examples take advantage of the ability to communicate spoken information by modulation of a pulse train. One technique is to drive a piezoelectric transducer at a very low frequency, well below its natural or designed resonant frequency. One of the following examples drives the transducer by modulating the transducer's supply voltage with the envelope of the spoken words. Doing so causes the amplitude of the transducer's output to track the verbal amplitude, giving an understandable representation of the spoken message. In addition, because of a typical “cut-off” characteristic wherein the transducer will produce no sound if the supply voltage fall below a certain threshold, some frequency content can be reproduced as well. To maximize the volume of the resulting speech, the source material should be optimized in frequency content, preferably near the transducer's resonant frequency. Likewise, speech intelligibility can be optimized by altering the pace and intonations in the source material. Piezoelectric transducers that are, for instance, 2 mm thick and 6 mm square can produce sound pressure levels in excess of 85 dB while consuming only about 20 mA of current. These levels are easily heard by bystanders. The small size permits inclusion of this piezoelectric “loudspeaker” in small, light packages that can be integrated into a body-worn monitoring system, and the inherent design of the piezoelectric transducer produces high audio volume with minimal power demand.

[0019] In another example, a frequency modulation technique is employed. Turning to FIG. 1 an example of a frequency modulated system is shown. Electrodes or sensors 10 are in contact with the skin of the patient to monitor a body sign such as the heartbeat or ECG. Signals received by these sensors are sampled and digitized by a multiplexer and analog to digital converter 12. The digitally sampled signals are coupled to a microprocessor 20 where they are processed and analyzed for indication of a medical condition such as cardiac arrest. The mux and ADC 12 is also coupled to receive signals produced by a piezoelectric motion sensor 14 which can detect, for instance, whether the patient has just experienced a sudden fall as would occur if the patient suddenly became unconscious. The sensor data is also coupled to the microprocessor 20 for processing and analysis. A memory 22 is coupled to the microprocessor 20 and will generally contain program data, store patient physiological data at the time of a medical emergency for instance, and will store one or more prerecorded voice responses in accordance with the present invention. If a serious medical condition is detected by the microprocessor the microprocessor will recall the appropriate response data from the memory 22. For instance, a period of no signal from the motion sensor (the patient is still, unconscious) accompanied by arrhythmic heart signals may be interpreted by the system as the onset of cardiac arrest. An output of the microprocessor 20 is coupled to the input of a digital to analog converter 24 which produces an analog output signal from the response data which is smoothed by a subsequent low pass filter 26. The digital to analog converter 24 can have whatever resolution the designer desires and can be as simple as a one-bit D/A converter. The filtered analog signal is applied to the input of an amplifier 30 which drives a piezoelectric transducer loudspeaker 32. In a constructed implementation of this example the amplifier 30 had a complementary drive output. This can be constructed as a push-pull output amplifier as illustrated by transistors 62,64 with output terminals 66,68 in FIG. 3a, or as an H-bridge output amplifier as illustrated in FIG. 3b, in which the transistors 62,64 are driving piezoelectric loudspeaker 32. The transistors 62,64 are driven by complementary signals as illustrated at the base electrodes of the transistors. The output terminals 66,68 of the amplifier are coupled to the electrodes of the piezoelectric transducer 32. The output amplifier can be powered over a range of voltages such as ±3 volts to ±12 volts. Other voltages ranges may also be acceptable in a given construction.

[0020] FIG. 2 is an example of the type of drive signal which may be produced for the piezoelectric transducer 32 in FIG. 1. This drawing illustrates a binary pulse train 50, the form of which can be stored in the memory 22. When the pulse train 50 is converted to an analog signal, filtered, and applied over a voltage range of ±V₃, a drive signal such as 52 results. As the drawing shows, cycles of the drive signal are in proportion to the widths of segments of the binary pulse train, which in effect is a pulse width modulated signal. The pulse width modulation results in a frequency modulated signal which is used to drive the piezoelectric transducer 32.

[0021] FIG. 1a illustrates a power subsystem which can be used to power the components of the system of FIG. 1. A rechargeable lithium ion battery 40 provides a battery voltage V_B such as 3.6 volts. A lithium ion battery has a relatively low internal series resistance which, in conjunction with the use of a piezoelectric transducer for the loudspeaker, will maintain its output voltage while current is drawn from the battery. The battery voltage V_B supplies power for a power supply 42 which provides the necessary voltages for the system. In one constructed implementation the battery produced four supply voltages: 1.8 volts (V₁) for the low power circuitry, 3 volts (V₂) for the memory which used a flash memory device, and ±3 volts for the piezoelectric transducer drive circuitry. As mentioned above, other voltages may be used depending upon the needs of the circuitry used in the design.

[0022] FIG. 4 illustrates another example of the present invention in which amplitude modulation is used for the audio signal. As previously mentioned, in such an example the piezoelectric transducer is driven by modulating the transducer's supply voltage. The digitized, prerecorded voice message is accessed from the memory 22 by the microprocessor 20 and applied to an n-bit D/A converter 124. The output voltage from the D/A converter 124 is smoothed by a lowpass filter 26 and applied to the reference voltage inputs of two voltage variable power supplies 34 and 36, which are complementarily powered by supply voltages +V₃ and −V₃. The outputs of the voltage variable power supplies 34 and 36 are coupled to the electrodes of the piezoelectric transducer 32, driving the transducer with a voltage that varies in the range of ±V₃. Some users may prefer the voice clarity of this amplitude modulated technique over the frequency modulated technique described above.

[0023] FIG. 5 is an exploded view of a body-worn sensor which is small, light, and attaches to the body by a conductive adhesive. The sensor has a clamshell case of two halves 82 and 84. On the outside of the case half 82 are electrodes coated with a conductive adhesive that attaches the sensor to the body and conducts vital sign signals to the electrodes. The electrical components of the sensor are located on a printed circuit assembly 80, including the piezoelectric motion sensor 14. The battery 40 is located between the printed circuit assembly and the case half 84. The piezoelectric transducer 32 may be located on the printed circuit assembly 80 as shown, or may be attached to case half 84 to take advantage of the acoustic properties of the case and better transmit the voice message to bystanders.