This invention relates to a system using movement and sound-based information to count coughs or evaluate othe discrete activities of a patient.

Background of the Invention Coughing, or the sudden, forcible expulsion cf air a high velocity, is invariably accompanied by sounds havin selected frequency characteristics and differing intensities Patients with persistent nocturnal coughs frequently complai of sleep disruption caused by the cough reflex, as do thos capable of hearing the oft-unpleasant sounds associated wit the coughs, including parents of acutely or chronically il children. To both the patient and those in close proximity therefore, suppression of the patient's cough reflex i frequently highly desirable. This desirability has led t experimentation with a wide variety of substances as antitus sive agents, with the experimentation itself causing a paralle effort to develop methods for accurately and uniformly deter mining the effectiveness of these medications.

Early attempts at artificially inducing coughing wit chemical irritants or using subjective responses in ascertain ing effectiveness have been largely unproductive and widel criticized. Artificially inducing coughs leads to increasingl inaccurate measurements as patients develop tolerances to th irritants, while varying the frequency and intensity of th

^S UBSTI

coughs as a function of the concentration of the irritant an the interval between exposures. Although subjective response may provide valuable evidence of patients' beliefs as t effectiveness, they lack readily quantifiable informatio useful for comparing the wide range of competing antitussives In addition, controlled studies have shown the potentiall grossly misleading nature of purely subjective responses.

More recent attempts at objectively determining th usefulness of antitussives have included tape recorders a means for capturing sound information concerning coughs emitte by patients. As discussed in Sevelius and Colmore, "Objectiv Assessment of Antitussive Agents in Patients with Chroni Cough," The Journal of Hew Drugs 216-23 (July-August 1966) sound information related to coughing can be transmitted to recorder using a throat microphone taped to a patient Evaluation of the antitussive includes counting the cough emitted by each patient, a process conducted by manuall reviewing th*> ra e recording for each patient. According t Sevelius and Colmore, although intrusive, the throat microphon and attached wiring does not prevent patients from watchin television moving relatively freely throughout a room. Such a intrusive system is, however, likely to disrupt normal sleepin patterns, decreasing both the comfort levels and movemen abilities of patients wearing the throat microphones. Other devices presently used for monitoring the onse of coughs often include a belt worn about the abdomen of th patient and sensors maintained in contact with the body Typical of these devices are those described in U.S. Paten Nos. 4,413,620 and 4,240,444 to Tucker and Virgulto, et al. , respec tively. The patent to Tucker, for example, discloses an ab dominal restraint system having a belt designed to inflate i response to signals associated with the onset of coughs Pressure sensors inserted into the patient using catheters o mounted within the belt are used to detect internal body motio associated with coughing. The apparatus disclosed in Virgulto et al. similarly includes a belt attached to the patient's bod

SUBSTITUTE SHEET

and uses a protruding, displaceable disc to contact a seale bulb and change its pressure relative to the atmosphere.

Although both the Tucker and Virgulto, et al. device respond to the onset of coughing, neither apparatus is designe to evaluate the effectiveness of cough-resisting medications b counting the number of coughs emitted. As with the system o Sevelius and Colmore, the intrusiveness of these devices addi tionally limits their use as evaluation tools in nocturna environments. Discomfort felt by patients while wearing th devices is likely to inhibit normal sleeping patterns, thereb decreasing the accuracy of any measurements sought to be made.

U.S. Patent No. 4,848,360 to Palsgard, et al., dis closes a less intrusive system used to prevent snoring. According to the Palsgard, et al. patent, a microphone place beneath the patient's mattress may be used to pass soun information to a series of high- and band-pass filters. Th filtered signals subsequently are sent to circuitry designed t measure their periodicity and determine whether the periods correspond to typical time intervals for snore-associated sounds. Once snoring is detected logical signals are trans¬ mitted to a counter which activates a vibrational device or other anti-snoring apparatus following receipt of a preselected number of signals.

Like the devices disclosed in the Tucker and Virgul- to, et al. patents, the system of Palsgard, et al. is not designed as an evaluation tool for the counting of coughs. The Palsgard, et al. system also lacks storage means beyond the digital counter used to determine if continuous snoring is occurring, preventing any off-line evaluation or visual confirmation of the sleeper's activities. Because some snoring may not alter respiration sufficiently to result in deter- minable movement by the patient, the system disclosed in Palsgard, et al. similarly lacks any motion detecting means

SUBSTITUTE

Summary of the Invention The cough-counting system of the present invention includes means for providing both movement- and sound-base information obtained from a patient to an analog (or digital) storage device such as a videocassette recorder (VCR) . Th unintrusiveness of the system, which includes a sensor place beneath the patient or the patient's mattress, avoids interfer¬ ing with normal sleeping or waking activities. At the sam time, however, audio and visual data derived from both the movements and sounds of the patient increase the accuracy of the cough count and provide means for confirming the count if desired. Storing the data also allows processing to be conducted at any convenient time, further reducing the likelihood of disturbing the patient's sleep.

One embodiment of the present invention includes pressure transducer, video camera, directional microphone, an

VCR having two audio channels. Movement of the patient alters the pressure of an unintrusive, partially inflated air mattress relative to atmospheric pressure, varying the voltage emitte by the associated transducer. The transducer signal then is encoded using modified delta-modulation techniques and signals obtained from the encoder and microphone are transmitted to the two audio channel inputs of the VCR. An infrared light sourc provides illumination unseen by the patient, allowing the patient's movement to be recorded using the video camera (which may be equipped with an auto diaphragm to extend its dynamic range) connected to the video channel input of the VCR. Part or all of the information stored using the VCR may be fed to signal processing equipment, and the audio track may be frequency filtered to determine the presence and number of coughs. Because the encoded information derived from the transducer frequently is sufficient to generate an accurat cough count, the other audio and video data may be used as confirmation if desired. Additional verification informatio may be obtained using passive techniques designed to recor variations in the patient's infrared energy emissions.

^S UBSTITUTESHEE _T

The present system also is capable of developing information relating to respiration and heart rate, functioning as a ballistocardiograph, and determining the presence of symptoms such as restlessness, sleep apnea, cardiac arrythmia and, in some cases, bed wetting. Each of these conditions manifests itself in differing patient sound, movement, and infrared emission characteristics, allowing the present system to determine its existence through appropriate processing of information such as that available at the audio and visual inputs of the VCR. By examining these conditions the invention may be used to determine the effectiveness of medicines such as sedatives and hypnotics.

It is therefore an object of the present invention to provide a system for counting the number of coughs emitted by a sleeping patient.

It is another object of the present invention to provide a system for evaluating the effectiveness of cough- resisting medications by counting the number of coughs emitted by a medicated patient. It is an additional object of the present invention to provide an unintrusive system for detecting both sounds emitted by and movement associated with a patient.

It is a further object of the present invention to provide a system using audio and visual information to evaluate the activities of a patient.

It is yet another object of the present invention to provide a system for recording movement- and sound-based audio and visual information for off-line evaluation of the informa¬ tion relating to coughs, heart rate, respiration, car- diodynamics, or other activities of a patient.

Other objects, features, and advantages of the present invention will become apparent with reference to the remainder of the written portion and the drawings of this application.

SUBSTITUTE S

Brief Description of the Drawings FIG. 1 is a partially schematic representation of th system of the present invention used to evaluate the activitie of a. sleeping patient:.

FIG. 2 is a schematic representation of an electroni circuit used to encode analog signals obtained in connectio with the system of FIG. 1 for subsequent processing.

FIG. 3 is a schematic representation of an electroni circuit used to decode the signals encoded using the circuit o FIG. 2.

Detailed Description of the Drawing FIG. 1 details the cough-counting system 10 of th present invention. System 10 includes a pressure transduce 14, video camera 18, microphone 22, and storage device 26 suc as a VCR, digital worm drive, optical recorder, or othe appropriate storage means. Also shown in FIG. 1 are sensor 30, placed beneath a conventional mattress 32 and connected t transducer 14 by connector 34, encoder 38, and infrared ligh source 42. System 10 is designed to count the number of cough emitted by a sleeping or waking patient 46, typically a youn child but including an adult or an animal, during a preselecte interval. By allowing evaluation of both movement- and sound based data derived from patient 46, system 10 provides a accurate and easily verifiable determination of the amount o coughing which has occurred.

In one embodiment of the present invention, system 1 includes as storage device 26 a VCR having two audio inpu channels 50 and 54 and a video input 58. Signals derived fro a gauge- or piezoelectric-type transducer 14 and a directiona microphone 22 are provided to audio inputs 50 and 54 respec tively, while information obtained from video camera 18, whic may include an auto-diaphragm or -iris to extend its dynami range, is directed to video input 58. Magnetic tape within th VCR may be used to record the data for subsequent evaluation, allowing off-line processing to be performed at a remot facility. As a result, system 10 may be conveniently use

^S UBST ^I TUTESHEET

inside the patient's home, reducing difficulties associated with monitoring patient 46 in a foreign environment.

Before patient 46 retires to bed 62, sensor 30, illustrated in FIG. 1 as an air mattress approximately 24" x 36" x 3" in dimension when fully inflated, is placed beneath conventional mattress 32 (or, if desired, beneath patient 46) and inflated to a depth of approximately 1". Partially inflating sensor 30 maintains the air mattress at approximately atmospheric pressure while minimizing the variation in mattress 32 firmness experienced by patient 46. At the same time, however, sensor 30 remains sensitive to pressure changes caused by, for example, the coughing, respiration, or heart rate of patient 46. Once patient 46 enters bed 62 a baseline pressure may be determined. The patient's surroundings also may be illuminated by infrared light source 42, allowing infrared recording video camera 18 to function in either the absence or presence of visible light.

Evidence of coughing emitted from patient 46 may be recorded in a variety of ways. As the diaphragm of patient 46 moves in response to a cough, pressure deviations at sensor 30 are transmitted to transducer 14. Hollow plastic tubing may be used as connector 34 to transmit air pressure information to transducer 14, although those having ordinary skill in the art will recognize that sensor 30 itself may be a pressure transducer or any other device suitable for measuring movement of patient 46. Transducer 14, in one embodiment having a range of 0-1.5 psi and capable of withstanding 15 psi without calibration shift, converts the differential pressures (rela¬ tive to atmospheric pressure) into electrical signals for storage on the magnetic tape contained within the VCR or other appropriate storage means used as storage device 26. Because the output of transducer 14 usually falls within a relatively narrow range of values and contains accuracy-reducing noise components, the signal may be passed through a low pass filter (not shown) before being encoded for recording.

Microphone 22 and video camera 18 similarly may record the presence of coughing, with the microphone 22

^S UB ^S T ^I TUTESHEE

receiving sound waves emitted by patient 46 and the video camera 18 recording the muscular spasms associated with coughs. Signals from microphone 22 are transmitted to audio input 54 of storage device 26, while the signals from video camera 18 are forwarded to video input 58. A passive infrared system 66, designed to detect variations in the infrared energy emissions of patient 46, may be used to determine the presence of coughing as well.

Encoder 38 (FIG. 2), designed to encode the analog signal from transducer 14 in serial form, includes a comparator

70, flip-flop 74, fixed delay generator 78, resistor ("R^"), and capacitor ("C _] _" ). An analog signal, which may be the output of transducer 14 and in some embodiments has a bandwidth of 0-2.5 Hz, is transmitted to the positive terminal ( "+" ) of comparator 70. The output of comparator 70 then is strobed into the input ("D") of flip-flop 74 with each clock pulse

("CLOCK"), and the output value ("Q") of flip-flop 74 is sent through fixed delay generator 78 back to the reset terminal

("RE") of flip-flop 74. The output value ("Q" ) of flip-flop 74 also serves as the output of encoder 38, and is fed back to the negative terminal ("-") of comparator 70 through resistor

("Rl" ) and capacitor ( "C^' ) . As illustrated in FIG. 2, the set terminal ("S") of flip-flop 74 is grounded. In addition, choosing 5kHz for the "CLOCK" frequency and 10Hz for RiC _^ (i.e. the product of "R^" multiplied by "C _j ") makes pulse recovery relatively easy during processing and reduces tracking error substantially.

Encoder 38 performs a modified form of delta modula¬ tion. Traditional delta modulation includes comparison of an input signal to the average value of the system output, with the result of the comparison being clocked into a flip-flop. The output of the flip-flop also serves as the system output, which is averaged using an RC circuit and transmitted back to the positive terminal of the comparator. The system output consists of a square wave having transitions synchronous with the clock pulses, with the input information encoded in the transitions which do and do not occur. Although traditional

^S UBSTITUTE SHEET

delta modulation serves as a useful technique for encoding analog signals in serial form, its large bandwidth require¬ ments, resolution limitations, and inability to compensate for output data recovery drift diminish its effectiveness in connec ion with system 10 of the present invention. By modifying traditional circuitry to, among other things, include fixed delay generator 78, however, encoder 38 may compensate for such output drift and allow the output ("Q") of flip-flop 74 to maintain fidelity for all values from D.C. to 2.5 Hzat the comparator 70 input.

An important feature of encoder 38 is that it allows processing to begin at any point in the data stream, as the recovered information will converge to the correct input values with a time constant of lOR^C^. FIG. 3 details a circuit 82 for recovering the input data, which includes fixed duration pulse generator 86 and a resistor ("R2") and capacitor ("C2") network. The output of circuit 82 then may be appropriately processed to determine the quantity of coughs emitted (or other quantifiable behavior performed) by patient 46 during a specified interval.

The foregoing is provided for purposes of illustra¬ tion, explanation, and description of embodiments of the present invention. Modifications and adaptations to these embodiments will be apparent to those of ordinary skill in the art and may be made without departing from the scope or spirit of the invention. In particular, system 10 may be used to detect and evaluate information relating to respiration, heart rate, and various other types of motion and activities of patient 46, and may function as a ballistocardiogram. Moreover, sensor 30 may comprise a variety of multi-chambered air mattresses, depending on the type of patient 46 behavior desired to be measured. Single chambered mattresses allow air to move freely within the total volume enclosed, permitting transducer 14 to provide an accurate reflection of deviations in the total load on sensor 30. If measurements of more localized activity, such as various types of patient motion (including, for example, propping on an elbow) are desired,

^S UBSTITUTESHE

however, the single-chambered air mattress could be replaced with one having a number of seams extending much of the length or width of the mattress, thereby dividing it into narrow channels. The seams would prevent rapid equilibrium within the mattress and allow local transient pressure to be sensed at the output pressure port. Similarly, if localized data sensitivity is desired in addition to overall load pressure, the multi- chambered mattress could be designed to include seams extending virtually its entire width or length (but not completely sealing any channel from the others), with pressure ports placed in each channel, and separate transducers employed to allow outputs to consist of individual measurements or "sum¬ mary" recordings based on, for example, weighted sum and difference measurements.

SUBSTITUTE SHEET