GB2240392A - Acoustic monitor for vital functions - Google Patents

Abstract

A monitor for vital functions, such as respiration and heartbeat, is described which, in its preferred form avoids physical contact with the monitored subject, and provides an alarm signal when the functions deviate from a steady rhythm. One or more tuned microphones associated with an acoustic reflector receive sound from a patient and signal carrying means transmits the received sounds to a computer which selects sounds arising from one or more repetitive vital functions to provide a record and/or an alarm signal when the function or functions deviate from their normal rate. In an alternative form, sound receiving means uses signal carrying means associated with the patient to transmit the received sounds to the computer. The computer is preferably adapted to respond to the frequency and/or harmonic changes in the sounds emitted during inhalation and exhalation of breath. The received sounds may be converted to a digitally coded form before being received by the computer means, e.g. a digital code having at least 32 bits. The received sounds are sampled at a rate of at least 2048 per second.

Description

MONITOR FOR VITAL FUNCTIONS This invention relates to a monitor for vital functions, such as respiration and heartbeat, which preferably avoids physical contact with the monitored subject, and provides an alarm signal when the functions deviate from a steady rhythm.

Various monitors have been proposed which sense the vital functions of a patient by means of electrodes attached to the body of a patient. Such invasive means of monitoring are unacceptable for infants and certain adults due to the effect of movement which can dislodge electrodes resulting in false signals. In the case of infants particularly attachment of electrodes gives rise to skin disorders and irritation.

A serious problem with infants is apnoea, a cessation of breathing which may be temporary or permanent if no remedial action is taken. If action is not rapid anoxia and death can follow an apnoea episode.

Two forms of automatic infant monitoring are presently available. One uses transthoracic impedance pneumography in which changes in the impedance are measured between two electrodes attached to the chest wall; the changes correspond with the respiratory rhythm. The method requires signal leads to be attached to the patient. A second method of monitoring measures the impedance across the area of a pressure sensitive mattress. This system monitors general movement and is not specific the respiratory or cardiac rhythms.

Neither method is considered a reliable method of monitoring respiratory rhythm as chest movements can continue during an apnoea episode when no air is entering the lungs.

According to the present invention there is provided apparatus to monitor vital functions having sound receiving means comprising one or more tuned microphones associated with an acoustic reflector to receive sound from a patient, signal carrying means to transmit the received sounds to computer means adapted to select sounds arising from one or more repetitive vital functions and to provide a record and/or an alarm signal when said function or functions deviates from its normal rate.

There is further provided apparatus to monitor vital functions having sound receiving means comprising one or more tuned microphones associated with a signal carrying means to transmit the received sounds to computer means adapted to select sounds arising from one or more repetitive vital functions and to provide a record and/or an alarm signal when said function or functions deviates from its normal rate.

The sound receiving means consists of one or more microphones located to receive the sounds emitted by a patient.

Preferably the frequency response of the microphones is restricted, mechanically or electrically, so that the sensitivity to sounds associated with a vital function, e.g. the resonant frequency of the larynx, is enhanced and the sensitivity to other sounds from the monitored patient and from the surroundings is reduced. Background sound reduction systems, such as the use of a supplementary microphone connected in antiphase, may be used but are seldom essential.

The signals from the sound receiving means are carried to a computer for signal analysis. The signals are preferably transmitted without the use of wires or fibres by electromagnetic radiation such as light, infra-red radiation or radio waves. The signals may be amplitude, frequency or phase modulated on the carrier if in analogue form.

Digital coding of the analogue audio signal prior to transmission allows the use of FSK, PCM or other modulation methods for encoded data to be used. Radio frequency signals in the VHF band, 70 to 250 MHz may be used.

Because of the bandwidths involved the carrier is preferably in the UHF, SHF or higher bands. Signal transmitting methods in these bands are well known. In most applications the transmission range between the sound detection means and the computer means will be from 2 to 250 metres.

Preferably the sound receiving means is directional so as to maximise sound signals from the patient and minimise background noise. A preferred device for providing directional properties is a parabolic reflector with a microphone at its focus. Noise cancelling systems may also be used to reduce extraneous sounds from the environment. However these sounds are usually rejected through the pattern recognition system used for signal analysis.

In an alternative embodiment the sound receiving means is located close to the patient so as to maximise sound signals and minimise background noise. However sounds transmitted by conduction from the patient must be avoided as these may interfere with the desired external sounds emitted by the patient comprising function rhythms modulated by the patient's larynx resonances. A preferred device for transmitting the received sounds to the computer means is a so-called 'radio microphone' consisting of one of more microphones, an amplifier and a low power radio transmitter emitting signals in the VHF or UHF band. The radio microphone may be an integral device within a small case with a short trailing aerial or the microphone section may be separate from the case and connected by a short lead.In either case the device may be clipped or otherwise attached to the patient's clothing to ensure close proximity to the source of the sounds to be monitored. It should be accoustically separate from the body to avoid internal resonances of the breathing rate which may interfere with the desired breathing rate signal.

The computer means is adapted to receive the audio signals from the sound receiving means and filter, select and store those rhythms associated with a vital function such as the respiratory rhythm. Pattern recognition from the signals is obtained by continuously matching with data stored in the computer. Continuous correlation of each cycle of the rhythm with previous cycles is performed and alarm means initiated whenever short term variations occur in the repetative functions. Long term variations caused by waking, sleeping and other variations in patient behaviour are accomodated by the computer control system without generating alarm signals. However cautionary signals may be generated to indicate incipient problems which may require minor alterations to the patient's management.

In the most preferred form of the invention the sound signals are converted to digital form and processed by the computer means in this form. In order to provide the necessary flow of information to the computer means the analogue signals provided by the microphone are converted to a digitally coded signal having at least 32 bits. The sampling rate is preferably at least 25 kilobits per second. Under software control the variations in the signals are compared with a mathematical template which emulates, or carries a 'fingerprint', of the frequencies produced from the respiratory tract of a patient in many conditions. Such information is available from certain hospital data banks and records. The patterns are compared with a large number of breathing patterns relating to normal and abnormal conditions which are held in a data bank.Pattern matching operations are preferably carried out at a rate of 2048 per second. Such a set of frequencies and patterns, a 'fingerprint', can be 'learnt' from the patient while in a state of well being and stored in memory. (Very long term changes in characteristics caused by growth in infants can be accomodted by the computer means and used to modify the 'fingerprint'.# The record of a patients 'fingerprint' may be stored on magentic media, as optical code or on ROM to allow transfer from one computer means to another. For example, the signals from a patient in hospital may be monitored on a large mainframe computer. Subsequent transfer to home or a nursing home may use a small microcomputer but use the pre-recorded fingerprint.

Both respiratory and cardiac rhythms may be extracted from the acoustic signals and used in conjunction to define a 'steady state' for the monitored patient. Medical staff may insert control data according to the nature of the patient's disability or anticipated problems. The computer operating system will take into account the control data and acoustic signals so as to provide an alarm signal when a rapid change takes place in one or more of the patients vital functions and also provide a record showing long term changes, if any, in of the patient's functions. By observing the long term changes medical staff will be able to intervene by chemotheraputic or other means to stabilise the patient. Such self-adaptive computer systems are often called 'Expert Systems'.

For the detection of apnoea the respiratory function is the most important characteristic of the patient to be monitor ed. The passage of air down any length of tubing, such as the trachea and primary bronchial tubes, is modulated by the resonant frequency or a harmonic of the cavity. The primary resonant frequency depends on the volume and shape of the organs, the composition of the gaseous mixture within them and its temperature. The input gaseous mixture will be primarily air, a mixture of oxygen and nitrogen, together with an amount of water vapour depending on the ambient relative humidity. The exhaled gaseous mixture will contain less oxygen than the input mixture, a significant amount of carbon dioxide and an increased amount of water vapour. There will normally be an increase in temperature and the exhaled gases will have a higher temperature than the input gases.As a result of the composition and temperature changes in the gases as they enter and leave the lungs there a will be a significcant change in the harmonic content of the sounds generated in the cavity. A frequency shift takes place in the resonant frequencies of the organs during inhalation and exhalation due to this difference in gas composition.

This shift will be further emphasised by changes in the vocal chords which are relaxed during inhalation but become tense during exhalation. Some of the frequencies involved may be outside the audio spectrum.

Detection of the frequency shift and the harmonic content change between inhalation and exhalation is an important part of the preferred pattern recognition process. The signals received from the microphone are processed in digital form so that the pattern of frequencies associated with the organs involved in respiration can be selected from the signals presented to the computer means. Initially the system will be set to the basic parameters of the patient who may be infant or adult. The self-adaptive software will compare the signal with the entered para meters and rapidly alter the latter to match the characteristics of the patient when stable. Deviations from this stable state, by an amount preset in the computer, will be recorded and/or used to generate appropriate alarm signals.

Any computer system can be used which has adequate storage and processing capacity to accept and process the audio signals from a patient. In general where one patient is being monitored a single microprocessor system such as a standard PC will have sufficient computing power to provide the necessary output. In a hospital or other environment where a number of patients are maintained within a building or building complex the audio signals may all be sent to a single mainframe computer where they can be multiplexed and processed to provide separate output alarm signals for each patient.

The alarm signal provided by the computer means may be directly associated with the means and comprise audio, visual or combined signals of an attention attracting nature. Alternatively the signals may be transmitted by radio, wire or other means to an area where nursing staff are present. Under hospital conditions it is likely that a large mainframe computer may be used to process the data from a number of patients with alarm and monitor outputs directed to the relevant nursing area. Under home conditions it is possible for landlines to be used to transmit the received signals from the patient to a large computer within a hospital or available as a service by the land line provider.

Under home nursing conditions the computer output warnings may be sent in known manner through the domestic mains wiring network to an indicator in another room. In either hospital or home environments the wraning and alarm signals may be transmitted to a personal receiver, a so-called bleeper', via a signal carrying loop or by direct radio transmission.

The monitor has been described with particular reference to the reception and analysis of audio signals emitted from a patient. The monitor may include temperature sensing means either using emission from the patient or in close association with the patient to provide further information for use by the computer system. Skin resistance and other physiological changes may be utilised to provide expert diagnostic and prognostic information.

The monitor has been described with particular reference to human patients. The monitor may also be used with nonhuman species and particularly with mammals of appropriate size such as cattle and horses.

Claims (6)

1. Apparatus to monitor vital functions having sound receiving means comprising one or more tuned microphones associated with an acoustic reflector to receive sound from a patient, signal carrying means to transmit the received sounds to computer means adapted to select sounds arising from one or more repetitive vital functions and to provide a record and/or an alarm signal when said function or functions deviates from its normal rate.

2. Apparatus to monitor vital functions having sound receiving means comprising one or more tuned microphones associated with a signal carrying means to transmit the received sounds to computer means adapted to select sounds arising from one or more repetitive vital functions and to provide a record and/or an alarm signal when said function or functions deviates from its normal rate.

3. Apparatus as claimed in either claim 1 or claim 2 in which the computer means is adapted to respond to the frequency and/or harmonic changes in the sounds emitted during inhalation and exhalation of breath.

4. Apparatus as claimed in any of the preceding claims in which the received sounds are converted to a digitally coded form before being received by the computer means.

5. Apparatus as claimed in claim 4 in which the received sounds are converted to a digital code having at least 32 bits.

6. Apparatus as claimed in claim 4 or claim 5 in which the received sounds are sampled at a rate of at least 2048 per second.