GB2089999A - Non-invasive haemodynamic performance assessment - Google Patents

Abstract

For non-invasively measuring and assessing patient haemodynamic performance, two sensor systems are used as the raw data inputs. The first sensor system is an ear densitometer 12 which photoelectrically records the instantaneous amount of blood within the blood vessels of the ear. The second sensor system is a standard electrocardiogram on lead 19. The densitometer and electrocardiogram signals are conditioned and processed to permit measurement of specific performance parameters. These are pre-ejection phase, PEP; left ventricular ejection time, LVET; and PEP/LVET. The measured quantities are displayed on a cathode ray tube 28 in a variety of formats upon operator request and recorded by a printer 30. <IMAGE>

Description

SPECIFICATION Non-invasive patient performance assessment The present invention relates generally to electronic medical devices and more specifically relates to devices for electronically monitoring patient performance, such as cardiac performance.

Many systems are currently available for measuring cardiac performance parameters. A typical device is described in U.S. Patent No. 3,815,583, issued to Scheidt. This device uses photoelectricity to measure pulse rate of a patient. The device is attached to the patient's ear. A more sophisticated photoelectric pulse rate monitor is disclosed by Page in U.S. Patent No. 3,858,574. An even more sophisticated monitoring system is taught by Nakayama in U.S. Patent No. 3,920,004. Nakayma teaches a system for measuring blood flow and blood pressure.

Significant parameters not measured by Scheidt, Page or Nakayama are called systolic time intervals (STI). Considerable work has been done in the use of STI for diagnostic monitoring purposes.

The three parameters thought most significant are pre-ejection phase (PEP), left ventricular ejection time (LVET), and the ratio PEP:LVET. PEP is the length of time from the beginning of ventricular depolarization to the opening of the aortic valve. LVET is the length of time the aortic valve is open.

Chapter 6 of Progress in Cardiology, Vol. 1, entitled "The Systolic Time Interval as a Measure of Left Ventricular Performance in Man," by Weissler, et al, published by Lea and Febiger, Philadelphia, Pennsylvania,1972, discusses the determination and uses of PEP, LVET and PEP/LVET in diagnosing cardiac dysfunctions. Weissler, et al, recommend determining STI by charting electrocardiogram, phonocardiogram and carotid arterial pulse signals. The method taught has great medical significance but has practical problems associated with the measurement and calculation process.

In an attempt to overcome the practical difficulties associated with the Weissler, et al system, Cormier in U.S. Patent No. 4,094,308 teaches an automated system for determining PEP, LVET and PEP/LVET using only the phonocardiogram input. Such a system is extremely complex because of the difficulties associated with reliably processing the acoustic inputs.

According to the invention, there is provided apparatus for determining a physiological parameter of a patient comprising: first means for generating a first signal indicative of quantity of blood in a portion of said patient; second means for generating a second signal indicative of electrical activity of an organ of said patient; and means responsively coupled to said first generating means and said second generating means for computing said physiological parameter using said first signal and said second signal.

The preferred embodiment of the invention is apparatus for measuring systolic time intervals in which said first generating means is a densitometer, e.g. an ear densitometer, and said second signal is an electrocardiogram.

The preferred embodiment of the invention overcomes the difficulties associated with processing the complex phonocardiogram because it does not use an acoustic input. The only inputs are the electrocardiogram (ECG) and densitometer. Using the two inputs, the apparatus calculates PEP, LVET and PEP/LVET digitally and displays the results in both graphic and tabular form. By using an ear densitometer, the timing errors associated with sensor placement of a carotid monitor have been greatly diminished. By providing automated setup, measurement, sensor correlation, and parameter calculation, most inaccuracies due to operator error are eliminated. The densitometer input is accurately and easily analyzed by analog differentiation and digital correlation and analysis of the signal and its derivative.The operator may easily select a graphical representation of the raw and processed data which includes both atrial and ventricular artificial pacing data if present.

An embodiment of the invention will now be described by way of example and with reference to the accompanying drawings in which: FIG. 1 is an overall schematic view of a system employing the present invention.

FIG. 2 is a front view of the electronic module of the preferred embodiment of the present invention.

FIG. 3 is a block diagram showing the major hardware elements of the electronic module.

FIG. 4 is a schematic diagram of the AMP/DIF 11 6.

FIG. 5 is an electrical schematic diagram of AMP 1 14.

FIG. 6 is an electrical schematic diagram of level control 134.

FIG. 7 is a schematic diagram of ECG controller 118.

FIG. 8 is a schematic diagram of printer control 140.

FIG. 9 shows the general format for the display menu allowing for operator viewing of the sensor inputs and for selection from the menu.

FIG. 10 is a highlevel flowchart of the main operating firmware program.

FIG. 11 is a flowchart of the main operating INTERRUPT firmware program.

FIG. 12 is a detailed flowchart of program MAIN.

FIG. 13 is a detailed flowchart of program SAMPLE$-INTERRUPT.

FIG. 14 is a detailed flowchart of program MENU.

FIG. 1 5 is a detailed flowchart of program WAITFORTOUCH.

FIG. 1 6 is a detailed flowchart of program ECGSERVICE.

FIG. 1 7 is a detailed flowchart of program DENSSERVICE.

FIG. 18 is a detailed flowchart of program PACFTNSERVICE.

FIG. 19 is a detailed flowchart of program ASSESSRUNSERVIGE.

FIG. 20 is a detailed flowchart of program HALT$MENU. uAMPLITUDES0FECG.

FIG. 21 is a detailed flowchart of program REFERENCE FIG. 22 is a detailed flowchart of program SAM$START.

FIG. 23 is a detailed flowchart of program ECG$DENSMEASUREMENT.

FIG. 24 is a detailed flowchart of program SAMSSTOP.

FIG. 25 is a detailed flowchart of program PARAMETERSCALCULATl0N.

FIG. 26 is a detailed flowchart of program PARAMETERS$EVALUATION.

FIG. 27 shows the format for the tabular display of processed data.

FIG. 28 shows the format for the tabular display of raw data.

FIG. 29 is a schematic diagram of beeper 113.

FIG. 30 is a schematic diagram of pacing spike detector 121.

The preferred mode of practising the present invention is described herein as implemented in a product of the applicants called the Non-Invasive Patient Performance Assessment (NIPPA) system.

This system uses a number of components which are commercially available. Therefore, a minimal description will be given concerning such commercially available components. The main attention given in this disclosure concerns the elements and components of this system which are not commercially available which are constructed expressly for practising the present invention.

FIG. 1 shows a schematic view of the overall use of the NIPPA system. As can be seen, the patient 10 has ECG leads connected to his right arm at reference point 14, his left leg at reference point 18, his right leg at reference point 17, and his left arm at reference point 1 6. In addition, an ear densitometer is connected at reference point 12. Many commercially available textbooks may be found which describe the connection of the ECG sensors. The above-referenced U.S. Patent No. 3,815,583, issued to Thomas L. Scheidt describes the installation and use of such an ear densitometer. The major discussion to be undertaken below is a description of the electronic module. Schematically, the major components of electronic module 20 are densitometer processing 24, ECG processing 22, timing control 26, display 28 and printer 30.Each one of these components will be discussed in a slightly different order within the body of this disclosure. What is to be seen in FIG. 1 is a means for connecting to patient 20 and the means for interfacing with the operator via display 28 and printer 30. Notice that no operator input device is readily apparent; however, as is discussed below, display 28 has provisions for operator entry of commands.

FIG. 2 is a front view of electronic module 20 of the NIPPA system showing the various features.

Display 28 is shown as the largest component on the front of the device. Cable 1 7 is the densitometer connection (see also FIG. 1). Cable 19 is the ECG cable (see also FIG. 1). Button 36 is used for overall system reset.

FIG. 3 is a block diagram of electronic module 20. The general purpose processor 112 is a Prolog Model 7801. It drives beeper 113 via cable 111. The display graphics processor 146 is four interconnected Matrox STD 256 devices. Display alphanumeric processor 148 is a Matrox STD 2480.

Video mixer 1 50 is described in the Matrox literature which accompanies the delivery of the display graphics processor 146 and display alphanumeric processor 148. The output of display graphics processor 146 arrives at video mixer 150 via cable 1 52. Similarly, the output of display alphanumeric processor 148 is transferred to video mixer 150 via cable 154. Video mixer 150 then supplies the combined video signal via cable 156 to CRT 158 and printer 144. CRT 158 is Ball Brothers Model TV90.

Printer 144 is Axiom Model EX850. Printer control 140 is connected to printer 144 via cable 142.

Printer control 140 controls the manual buttons associated with printer 144 which allow feeding of paper, starting, stopping, printing etc. Printer control 140 therefore is the interface between the digital signals received from 1/O 108 via cable 130 and the manual controls of printer 144.

The main digital distribution facility within electronic module 20 is the STD digital bus 110. STD or standard digital bus is a common protocol that is used throughout the industry. The various digital elements which are connected to STD digital bus 110 are commercially available with the STD digital bus protocol and are simply connected as can be seen in FIG. 3. These devices include USART 104, general purpose processor 112, memory 106, 1/O 108, display graphics processor 146, A/D converter 132 and display alphanumeric processor 148. STD digital bus 110 enables intercommunication with all these elements.

USART 104 is a specialized V6 device which is used to provide an interface between RS-232C interface cable 102 and STD digital bus 110. USART 104 is thus a universal synchronous-asynchronous receiver-transmitter common in the industry. Typical of such a device is Intel Device Part No. 8251A.

The operator input 100 is a Carroll Manufacturing Co. touch input kit. This device is a matrix of photoelectric devices and illuminating devices in a framework which is attached to the front of CRT 1 58.

The special purpose logic of operator input 100 senses a break in the transmission of light within this matrix and notifies USART 104 via cable 1 02 of the position of the interrupted light beam. This interruption signifies a touch of CRT 1 58 by an operator. The touch is made in response to an inquiry to the operator on CRT 158 and is in the form of push buttons which are graphically drawn upon the face of CRT 158 in the manner described below.

Memory 106 contains a ROM portion of erasable programmable READ ONLY MEMORY circuits providing storage for the programs necessary to operate the NIPPA system. The RAM portion of memory 106 is Prolog Model 7701 which provides temporary storage. l/O 108 is Prolog Model 7602 which gives eight input/output channels which are program addressable by general purpose processor 11 2. The input/output channels of l/O 108 are connected as shown.

These channels supply control signals to printer control 140 via cable 130 as explained above.

Additional channels of 110 108 convey control information to AMP 114 via cable 122, to AMP/DIF 11 6 via cable 124, to ECG controller 118 via cable 126, and to level control 134 via cable 128.

The densitometer inputs are received via cable 17 as discussed above. The raw analog information is sent to AMP 114 which amplifies the densitometer information and to AMP/DIF 11 6 which differentiates the densitometer information and supplies the amplified derivative via cable 11 9. Control cables 122 and 124 permit control of the gain of AMP 114 and AMP/DIF 11 6 respectively. The amplified analog information in the form of densitometer output is supplied to A/D converter 132 via cable 11 7. The amplified derivative information is supplied to A/D converter 132 via cable 11 9.

The normal ECG input is received via cable 19. ECG processing 120 is the electronics from Hewlett-Packard Model 1505A ECG System. ECG controller 11 8 is a special purpose element which supplies control information to ECG processing 120 in response to control information received from 110 108 via cable 126. ECG controller 11 8 thereby can simulate operator inputs which ECG processing 120 would normally receive via manual switches. These operator inputs control the mode of ECG processing 120. They control the lead selection, initial gain, and other related inputs. The processed analog ECG data is sent by cable 136 to level control 134.Level control 134 responds to control information received via cable 128 from l/O 108 to control the level of ECG information which is sent via cable 138 to A/D converter 132. Conductors 19a, b and c are also coupled to Pacing Spike Detector 121 which locates pacing spikes and notifies General Purpose Processor 112 via cable 123.

A/D converter 132 is an Analog Devices Model RTI--1225 which receives its control information via STD digital bus 110, supplies its digital output via STD digital bus 11 0, and receives its analog inputs as shown.

Using the hardware elements described, the programs stored in the ROM portion of memory 106 in conjunction with the RAM portion of memory 106 and general purpose processor 112 respond to operator inputs via operator input 100 and supply outputs to printer 144 and CRT 158. The analog data is received as densitometer input via cable 17 and ECG input cable 19. These signals are converted via A/D converter 132 and transferred to memory 106 via STD digital bus 110. All of the elements shown are deemed to be standard in the art and readily available commercially except AMP 114, AMP/DIF 11 6, ECG controller 11 8, level control 134, beeper 113, pacing spike detector 121 , and printer control 140. These elements are described in further detail below.

FIG. 4 is a detailed electrical schematic diagram of AMP/DIF 11 6. As can be seen, the analog densitometer information is received via cable 17 from a phone jack having a tip contact 200, ring contact 202 and sleeve contact 204. Sleeve contact 204 is grounded as shown and tip contact 200 and ring contact 202 are used to present an input voltage across resistor 207 which is on the order 12K ohms. The bias voltage is supplied as shown, which is on the order of +5 volts. This voltage is supplied through diodes 206 and 208 as shown, which reduce the supply voltage and thereby prevent patient discomfort from the light source.

The received raw analog voltage developed across load resistor 207 is supplied to AMP 114 via line 210. Capacitor 212 decouples the input, being on the order of .33 microfarads. Capacitor 214, being .1 microfarad couples the signal which is dropped across resistor 216, which is on the order of 1 00k ohm. The input to operational amplifier 224 is via resistor 218 which is also 1 00k ohm. The positive input of operational amplifier 224 is resistor 226 which is 51 k ohm.

The feedback network shown contains 1 M ohm resistor 22C and capacitor 222 which is .027 microfarads. The output of operational amplifier 224 is coupled to operational amplifier 232 via 50k ohm resistor 228. The other input to operational amplifier 232 is via 47k ohm resistor 230. Feedback resistor 231 is 1 meg ohm. The purpose of operational amplifier 224 and operational amplifier 232 is to supply the gain required by the other circuitry along with differentiating the input via the resistor/capacitor network shown. Gain control 234 is a commonly available part. The three bit digital input is via pins 4, 5 and 6 as shown. This three bit input is used to control the gain on the analog signal coupled via 4.7 microfarad capacitor 233 to pin 15. In this case, the analog signal is the amplified derivative of the densitometer signal. The analog output of gain control 234 is via pins 1, 2 and 1 6. Pins 2 and 3 are grounded as shown. The three bit gain control signal is received from l/O 108 via cable 124.

The output of gain control 234 is amplified by operational amplifier 240. Zener diode 238 clamps the output of gain control 234. Feedback resistor 242 is on the order of 1 Ok ohm.

Variable resistor 244 permits manual attentuation of the signal during initial setup of a particular unit. This is to be a maintenance type feature. Variable resistor 244 is on the order of 1 Ok ohm as well.

1 OK ohm resistor 254 couples the signal to operational amplifier 258. The positive input of operational amplifier 258 is connected to ground via 8,200 ohm resistor 256. The bias circuit for operational amplifier 258 contains resistor 248 which is on the order to 510k ohm. Potentiometer 246 is 50k ohm which is connected between +6 volts and6 volts. Feedback resistor 250 is 200k ohm with variable resistor 252 of 500k ohm in series. Again, variable resistor 252 is to be used for setup and maintenance purposes. The analog output of AMP/DIF 11 6 is transferred to A/D converter 1 32 via line 11 9 as shown.

FIG. 5 is a detailed schematic diagram of AMP 114. The input is received via cable 210 as explained above. AMP 114 has a function similar to AMP/DIF 11 6 except that it amplifies and controls the gain of the densitometer signal itself rather than the derivative of the densitometer signal. The main active elements of AMP 114 are operational amplifiers 306,312,328 and 340. Gain control 322 serves functions analogous to gain control 234 (see also FIG. 4). The analog densitometer signal is coupled to the negative input of operational amplifier 306 through 1.0 microfarad capacitor 300 and 130k ohm resistor 302. Feedback from the output of operational amplifier 306 is via 2.7M ohm resistor 304. The positive input of operational amplifier 306 is grounded via 120k ohm resistor 308.

The output of operational amplifier 306 is coupled to the negative input of operational amplifier 312 via 51 k ohm resistor 310. The 39k ohm resistor 314 couples the positive input of operational amplifier 312 to ground. Operational amplifier 312 has a feedback network comprising resistor 316 which is on the order of 500k ohm and capacitor 318 which is on the order of .002 microfarads. The output of operational amplifier 312 is coupled to gain control 322 via 1 Ok ohm resistor 320.

Zener diode 326 clamps the output of GAIN CONTROL 322 as shown. Operational amplifier 328 receives the output of gain control 322 and couples it to operational amplifier 340 via resistor 332 which is 20k ohm. Resistor 330, providing a feedback path as shown, is 20k ohm. The positive input of operational amplifier 340 is via 18 k ohm resistor 342. Operational amplifier 340 is biased via potentiometer 334 which is 50k ohm via resistor 336 which is 10 k ohm. The offset adjusted by potentiometer 334 is intended to be a factory or maintenance adjustment. Feedback for operational amplifier 340 is via fixed resistor 338 which is 200k ohm and variable resistor 344 which is 500 k ohm The frequency response of operational amplifier 340 is adjusted by switches 348, 352, 356 and 360 as shown.The capacitors which are selectively switched in by the switches provide this variation.

These capacitors are 346, 350, 354 and 358. The output of the AMP 114 is via line 117 which is transferred to A/D converter 132.

FIG. 6 is an electrical schematic diagram for level control 134. The purpose of level control 134 is to allow selection of ECG gain via cable 128 from l/O 108. This allows the program to adjust the gain of the ECG signal. As is shown in FIG. 6, three separate independent channels are present. The three channels are totally independent except that the digital gain control input is the same for each of the three channels. Similarly, the exact circuitry within each channel is identical. For that reason, only ECG channel 1 is shown. ECG channel 2 level control 442 is identical to it, as is ECG 3 level control 444.

The analog ECG data is received via cable 136 from ECG processing 120. ECG I is coupled via resistor 400 to operational amplifier 404. Resistor 400 is 100k ohm. The positive input to operational amplifier 404 is via 51 k ohm resistor 406. Resistor 402 is 1 00k ohm and supplies the feedback as shown. Filter 408 is a 50 or 60 Hz notch filter, depending upon the primary power source. This filter is required, since the ECG signal will be used after digitization by general purpose processor 112, and not to drive a normal strip chart recorder which has the effect of attenuating such noise. Resistor 410 is of the value 100k ohm and couples the signal to operational amplifier 414. The fixed 8k ohm resistor 412 couples the positive input of operational amplifier 414 to ground.The feedback network for operational amplifier 414 consists of 220k ohm resistor 416 and .01 microfarad capacitor 418. The output of operational amplifier 414 is coupled to gain control 424 via 1 Ok ohm resistor 420.

Gain control 424 is a typical commercially available digitally controlled analog gain control device having input at pin 15 and output at pins 1,2 and 3. The controlled digital control input is via pins 4 and 5 as shown. The gain control signal is received from l/O 108 via cable 128. Notice that the same two bit gain control signal is utilized by each of the three ECG channels. The output of gain control 424 is clamped by zener diode 426. The clamped output is supplied to operational amplifier 428 as shown.

Feedback resistor 430 has a value of 20k ohm.

The output of operational amplifier 428 is coupled via 20k ohm resistor 432 to operational amplifier 436. The positive input of operational amplifier 436 is coupled to ground via 1 8k ohm resistor 434. The feedback network for operational amplifier 436 consists of resistor 438 having a value of 82k ohm and an optional capacitor 440.

As explained above, the ECG 2 level control 442 and ECG 3 level control 444 are identical in construction. The output of level control 134 is transferred via cable 138 to A/D converter 132.

FIG. 7 is a diagram showing the construction and operation of ECG controller 11 8. As explained above, ECG processing 120 is Hewlett-Packard Model 1505A which is manually operated device. To permit it to be operated under control of general purpose processor 112, ECG controller 11 8 is employed. The actual control signals as received by ECG controller 11 8 are received from l/O 108 via cable 126. It is a function of ECG controller 11 8 to convert those signals into signals which simulate manual actions.

The AUTO-INSTO is a signal which clears the digital circuitry of the ECG processing 120 during lead change and other types of reconfigurations. The signal is received via cable 1 26 from l/O 108 and is inverted by inverter 450 before being applied to the AUTO-INSTO input of ECG processing 120. The push button test input to ECG processing 120 is via relay 454. A relay is used to sink the amount of current required. The coil of relay 454 is driven by a gated astable multivibrator 452 having a one second period. This gated astable multivibrator provides 1/2 second alternating action simulating manual engagement of the 1 millivolt calibrate push button.Decoder 456 is a 1 of 8 decoder which decodes a three bit input from l/O 108 to provide sensitivity level controls X1/2v, X1 and X2. These sensitivity controls correspond to manual buttons normally actuated by the operator.

In similar fashion, decoder 458 is a 1 of 8 decoder which decodes a 3 bit input from l/O 108 to provide signals for lead change and calibration. Notice that the outputs are 1-2-3 state, AVR-AVL-AVF state, V1-V2-V3 state, V4-V5-V6 state, special state and auto-calibrate state.

These are the lead selection options which would normally be made manually by the operator. The last item required of ECG controller 11 8 is to supply power to ECG processing 120. This is because in its normal configuration, Hewlett-Packard Model 1505A has its own power supply. Power regulator 460 is common in the industry and provides this function.

FIG. 8 shows a schematic diagram of printer control 140. Similar to ECG processing 120, printer 144 is intended to be a manually operated device. Therefore, it has manual controls which would normally be actuated by an operator. To allow these controls to be operated by general purpose processor 112 via l/O 108 and via cable 130, printer control 140 simulates the manual inputs. In this case, the problem is one of sinking enough current for the manual control inputs. The four inputs that are required are shown. They are: resolution, image reverse (allows black on white and white on black printing), paper feed, and print command which simply starts the printer in operation. Each of these is supplied by a relay which connects that input to ground. The relays employed are relay 470, relay 472, relay 474 and relay 476.The coils are driven by successive conductors of cable 130 which are coupled to 110 108.

FIG. 29 is a schematic diagram of beeper 113. It is the function of beeper 113 to produce a sound notifying the operator that an entry has been received. As can be seen in FIG. 3, beeper 11 3 is controlled directly by general purpose processor 112 via discrete line 111. Referring again to FIG. 29, 1/2 second one-shot 170 supplies a 1/2 second pulse from output Q in response to an input pulse received from general purpose processor 112 via line 111. Part Number 74C221 is typically used for 1/2 second one-shot 170 along with resistor 172 and capacitor 174 which have a 1/2 second time constant.

Gated Astable 180 is typically an RCA 4047. It operates at 2.2kHz as determined by capacitor 176 and resistor 178. The inverted output is coupled via 1 Ok ohm resistor 182 to transistor 184.

Transistor 184 is a 2N3700 having the collector coupled to a Panasonic Piezoceramic audio generator 186. Resistor 188 is 1 k ohm. It couples generator 186 to the positive voltage supply.

FIG. 30 is a schematic diagram of pacing spike detector 121. Whereas it would be possible for general purpose processor 112 to locate pacing spikes under software control, considerable efficiencies result from the use of special purpose hardware. Pacing spike detector 121 receives the ECG signals from lines 19a, 19b and 1 9c of cable 19 as shown. Referring again to FIG. 1, it is seen that this corresponds to right arm, left leg, and right leg respectively. It is thought that these points provide the most reliable pacing artifact signals.

Pulse former 260 of type 180297 receives the ECG signals via feedthrough capacitors 280.

Capacitor 278 has a value of .0039 microfarad. Capacitors 274, 276, 282, 284, 286, 288 and 292 are .68,. 1, .015,2.2,. 1, 6.8 and .022 microfarad, respectively. Resistors 290 and 294 are 1 Ok ohm and 240k ohm. These components supply pulse shaping parameters in accordance with the manufacturer's instructions.

Operational amplifier 262 rectifies the input received. Resistor 261 is 47k ohm. Field effect device 264 is a 2N5669 which thresholds the signal. Diode 263 couples to ground. Transistor 266 is a Hewlett-Packard 1351 coupled via a photoelectric link as shown and biassed by diode 265 and resistor 267. The digital output is lengthened by one-shot 268 which is a 4098. The time constant is determined by .22 microfarad capacitor 271 and 82k ohm resistor 269 to be about 5 milliseconds.

Resistor 275 is 4.7k ohm which couples the 5 millisecond pulse to transistor 270. The output of type 2N3700 transistor 270 is used to interrupt general purpose processor 112 via line 123 (see also FIG.

3). Resistor 273 has a value of 47k ohm.

Referring again to FIG. 3, it can be seen that the hardware just described is integrated within electronic module 20. Notice that control of the entire system is always via general purpose processor 11 2 which supplies control signals both via STD digital bus 110 and via the various ports of l/O 108.

The operator input is all via operator 100 which is the touch control system described.

The control scheme used can best be observed by reference to FIG. 9. Shown in FIG. 9 is the general format for menus used for operator control. Notice that the menu description 520 and push buttons 526, 522 and 524 are graphically drawn on the face of display 28. The actual operator command is received by the system by communication from the touch input kit as described above.

Menu description 520 contains the basic question or description of the operator options. As can be seen in Table A, Table A also shows the purpose of each of the ten menus and the inscription which would be displayed on each of the ten buttons Select 1 through Select 10. Reference to FIG. 9 shows that the ten buttons referenced 526 may all be used or only a portion thereof may be used. The two right-most buttons seen on FIG. 9 are button 522 and button 524. These are system level commands which are constant for every menu. They are UNFREEZE TRACES which is button 522, and button 524 which is SYSTEM INTERRUPT.

The upper half of display 28 shown in FIG. 9 is the analog information as received and digitized.

There are three ECG signals shown, that is signal 500, signal 502 and signal 504. In addition to that, densitometer signal 506 and derivative of densitometer signal 508 are displayed. The captions for these analog signals are found at the left-hand margin of display 28. The ECG designations, that is ECG II 510, ECG lil 512 and ECG 1514 are changed depending upon the leads selected. The captions used for the densitometer signal, that is DENS 51 6, and the derivative of the densitometer signal, that is DERIV 51 8, are constant. The general format of display 28 thus shown in FIG. 9 is representative of most menus to be used by the NIPPA system. Exceptions to this general format are the output formats which will be described in greater detail below.

TABLE A

SELECT SELECT SELECT SELECT SELECT MENU DESCRIPTION PURPOSE 1 2 3 4 5 1) SELECT DESIRED MAIN MENU ECG EAP DENS PACER FTN ASSESSMENT INSTO PAOCEDURE SETUP SETUP TEST RUN 2) ECE SET PROGRAM ADUUST SELECT SELECT SELECT SELECT SELECT ADJUST AS NEEDED ECG I-II-III AVR-AVL- V1-V2-V3 V4-V5-V8 SPECIAL AVF 3) DENSITOMETER ADJUST DENSITO- DENSITO- DERIVATIVE DERIVATIVE CONTINUE SETUP PAOGRAM DENSITO- METER METER SIZE INC SIZE INC PROGRAM ADJUST AS NEEDED METER SIZE INC SIZE INC 4) SELECT DESIRED PREPARE BLACK ON WHITE ON NORMAL DOUBLE PAPER FORMAT PRINTER WHITE BLACK WIDTH MOTH ADVANCE 5) SELECT NUMBER HOW MUCH 2 5 10 20 50 OF R-R INTERVALS DATA TO DESIRED. ASSESS 6) SELECT CHANNEL WHICH TOP SECOND COMPOSITE - FOR P-2AVE CHANNEL CHANNEL CHANNEL DETECTION FOR P-WAVE 7) PRESS START TO BEGIN START - - - RUN ASSESSMENT 8) AUN IN PROGRESS NOTIFIES STOP RUN - - - NUMBER OF INTER- OF PROGRESS VALS DETECTED = -----9) RUN COMPLETED SELECTION TABULAR GRAPHED RAW - SELECT DISPLAY OF OUTPUT DATA DATA DISPLAY PROGRAM FORMAT 10)RUN INTERRUPTED RECOVER DISPLAY ADJUST ADJUST CONTINUE ASSE8S SELECT ACTION FROM INTER- RESULTS ECG DENSITO- RUN RUN DESIRED RUPTED RUN METER TABLE A (Continued)

SELECT SELECT SELECT SELECT SELECT MENU DESCRIPTION PURPOSE 6 7 8 9 10 1) SELECT DESIRED MAIN MENU ANALOG OPTIONS PRINT STANDARD PAPER PROCEDURE OUTPUT DATA SET ADVANCE 2) ECG SET PROGRAM ADJUST INSTO ROTA TE GAIN GAIN CONTINUE ADJUST AS NEEDED ECG TRA CES INCREASE DECREASE PROGRAM 3) DENSITOMETER ADJUST - - - - SETUP PROGRAM DENSITO ADJUST AS NEEDED METER 4) SELECT DESIRED PREPARE PRINT CONTINUE - - FORMAT PRINTER PROGRAM 5) SELECT NUMBER HOW MUCH 100 180 240 - OF R-R INTERVALS DATA TO DESIRED ASSESS 6) SELECT CHANNEL WHICH - - - - FOR P-WAVE CHANNEL DETECTION FOR P-WAVE 7) PRESS START TO BEGIN - - - - RUN ASSESSMENT 8) RUN IN PROGRESS NOTIFIES - - - - NUMBGER OF INTER- OF PROGRESS VALS DETECTED = -------9) RUN COMPLETED SELECTION - - - - SELECT DISPLAY OF OUTPUT PROGRAM FORMAT 10) RUN INTERRUPTED RECOVER TERMINATE DISPLAY PRINT STANDARD INSTO SELECT ACTION FROM INTER-RUN RAW DATA DATA SET DESIRED RUPTED RUN FIG. 10 is a very high level flowchart of the main processing program. The beginning of operation after power up or system reset is a start at element 600. The various variables are initialized at element 602 and a mode is selected at element 604. The small loop shown at element 604 is to ensure that a mode is selected.Once selected, the proper processing program is executed at element 606 and when complete, control returns back to the mode select element 604 for selection of another mode. Notice that this is the general form for the major computations performed by the NIPPA system.

FIG. 11, on the other hand, shows the operation of the interrupt processing. Interrupt element 608 is the 7.5 level interrupt of general purpose processor 112. It is entered every four msec by an internally generated INTERRUPT signal. The element 610 increments a timer which keeps track of the timing for the rest of the programs in the NIPPA system. The basic four msec interrupt rate establishes the overall time limits for the system. Element 612 determines whether an A/D conversion is required. Notice that this decision is made every four msec since this portion of the program is traversed with each and every interrupt. If an A/D conversion is required, the input is done at element 614. The input multiplexer to A/D converter 132 is incremented at element 616 and element 618 determines whether all inputs have been received.Implementation of the input multiplexer of A/D converter 132 allows the transition from ECG I to ECG II to ECG Ill to the densitometer and to densitometer derivative signals. Five signals then are input at each four msec interrupt. After it is determined by element 618 that four sets of data have been received, the analog information on display 28 is updated at element 620 (see also FIG. 9).

If no A/D conversion is determined to be required at element 612, the printer processing is accomplished at element 622 and some additional buffer cleanup is done at 624.

Element 626 determines whether or not operator command processing is required. By referring again to FIG. 3 it can be seen that all operator commands are received from USART 104 and placed on STD digital bus 11 0. USART 104 receives the operator commands in serial form in RS-232C format from operator input 100. By consulting the Carroll Manufacturing Co. information on the touch input kit, it is determined that a given touch command is transmitted as four separate bytes over cable 102. Each time USART 104 receives one byte, element 626 on FIG. 11 determines that an operator command, that is to say, one byte of an operator command, is available and input is performed at element 628. The evaluation of those commands is done within the main program as shown in FIG. 10. The exceptions to this are the two right hand buttons shown on display 28 in FIG. 9 which are system level commands.

After the interrupt processing is completed as shown in FIG. 1 exit is made at element 630.

Having thus described the overall structure of the NIPPA system, the discussion will now turn to an indepth explanation of the individual program structures. The NIPPA firmware is divided into ten modules, as shown in Table B. Each module has a number of separate individual programs. Many of these programs perform very simple housekeeping functions and therefore will not be described herein in great detail. However, a number of programs are relatively complex and important to the operation of the NIPPA system, so they will be described in detail herein.

TABLE B 1: PROCTO MODULE EGDEN MODULE (cont.) 1) WAITFORTOUCH 7)UN$ROTATE 2) ERASE 8) ECQSERVICE 3) ERASE BOAROS 9) DEN$ADJUST 4) RIGHTSUTTONS 10) D$DEN$ADJUST 5) NOTWARITTEN 11) DEN$SERVICE 6) PRINT$ ADJUST 7) TWEL VESQUARES 8) TWOSQUARES 7 MAIN MODULE 9) WRITE 1) OUTPUTSERVI CE 2) USART$READ 2. MATH MODULE 3) UPDATE 4) SAMPLEINTERRUPT 1) PARAMETER CALCULATION 5) MAINPROGRAM 2) PARAMETER EVALUATION 8. IN FIN MODULE 3. DISPLAY MODULE 1) GRAPH$DISPLAY 1) BUMJOUT 2) PACTINSERVICE 2) TABLE$DISPLAY 3) RUN$START 3) DUMPSRAW$ DATA 4) RESTART 4) RAW$DATA$DISPLAY 5) SYSTEMTESTSERVICE 6) ASCI$LOAD 4. POROD MODULE 1) DATA$ACQUISITION 9.RSENSE MODULE 2) SPIKE$DATA 1) DELTA 3) REFERENCESA.MPLITUDE$OF$ECG 2) UPDATE (Looai) 4) FIND$NEXT$A 3) INITIALIZE$R$WAVE$SENSING 5) SAM$START 4) SAMPLE$ECG 6) SAM$STOP 5) NEW$DETECT$PTR 7) R$PEAK$LOCATION 6) ADVANCED$PTR 8) Q$PEAK$LOCATION 7) SEGMENT$LENGTH 9) P$PEAK$LOCATION 8)USE$SEGMENT 10) D$DEN$MIN$MAX 9)REFINE$RR 11) D$DEN$MIN$LOCATION 10)PROCESS$SEGMENTS 12) ECG$DEN$MEASUREMENT 11) PROCESS$CANDIDATE 12) QRS$DETECTED 13) CALIBBATE 5. MENRUN MODULE 1) PRINTSTART 10. LONG MODULE 2) NO$D$PRINT$STA RT 3) HALT$MENU 1) LONG$TABLE 4) OPTION$SERVICE 5) MENU 6) DATA$SET$SERVICE 7) ASSESS RUNSERVI CE 6. ECGDEN MODULE 1) ONINSTO 2) OFFINSTO 3) ADJUST 4) GAIN FREEZE 6) CHANGEGAIN 6) ROTA TE$STEXT FIG. 12 shows a flowchart for the MAIN program 632.MAIN program 632 is located within module MAIN, as can be seen on Table B. Upon entry of the MAIN program 632, the various variables are initialized at element 634. A procedure call is made at element 636 to procedure ERASEBOARDS which clears the display. A procedure call is made at element 638 to procedure TWOSQUARES which draws the two rightmost buttons. As explained earlier, these buttons have special significance as they are system level commands which are constant for all menus. A procedure call is made at 640 to the procedure RIGHTBUTTONS by which the two rightmost buttons are labeled. By reference back to FIG. 9, it will be remembered that these two buttons are button 522, which is labeled UNFREEZE TRACES, and the other is button 524 which is labeled INTERRUPT. A procedure call is again made at element 642 to procedure TWELVESQUARES which draws the remaining buttons.Procedure MENU is called at element 644 which effectively transfers control from the MAIN program to PROGRAM MENU. Exit 646 is shown in dashed lines as the control is transferred to procedure MENU and control should not again return to MAIN program 632 for exit.

FIG. 13 is a detailed flowchart of program SAMPLEINTERRUPT 648. This program is also located within module MAIN. SAMPLE$lNTERRUPT 648 is the MAIN INTERRUPT program described in FIG. 11.

FIG.13, however, describes the program in more detail. Element 650 increments the printer timeout counter. This is a counter which keeps time in increments of four msec since the incrementation occurs for each and every INTERRUPT and each INTERRUPT occurs at a four msec interval. Element 652 checks a variable to determine whether or not the CRT is frozen. If it is not, element 654 inputs from A/D converter, and element 656 increments the multiplexer to the next of the five possible inputs.

Element 658 determines whether all five inputs have been converted and after all five have been converted, element 659 checks if this is the fourth SAMPLE$lNTERRUPT. If it is, it calls procedure UPDATE which updates the analog information on display 28. Four successive samples for each trace are averaged by procedure UPDATE, so that display 28 is updated every 16 msec.

Should element 652 determine that the CRT is frozen, this signifies that processing is now being done or there is a printer dump of the display screen in process. Element 664 determines whether or not there is a dump in process. If not, element 673 merely cleans up the buffer. If there is a dump in process, element 666 determines whether or not the data output is complete. If the output is complete, element 670 terminates the output transfer. If element 666 determines that the data transfer is not complete, element 668 continues the output.

After the printer output determination is made, element 674 determines whether the buffer clean up has been accomplished. If it has, element 676 calls procedure PRINT$ADJUST which advances the paper. This is followed by a call to procedure MENU by element 678. Notice that the continuation of element 678 is shown in a dashed line. This is because element 678 calls program MENU and actually transfers control to program MENU and it is not anticipated that an exit of SAMPLES$lNTERRUPT will be made in normal fashion. The procedure call to procedure MENU at element 678 signifies that an operation has been completed and a new operator command is required. Should element 674 determine that the buffer cleanup has not been completed, element 680 determines whether or not an operator command byte is present in USART 104.If one is found to be there, element 682 calls procedure USART$READ which inputs the byte. Exit is made at element 662.

FIG. 14 is a detailed flowchart for program MENU 684. This program is located in module MENRUN as shown in Table B. Procedure MENU 684 is the program which controls the MAIN MENU as shown as the first entry in Table A. Notice from Table A this is described as SELECT DESIRED PROCEDURE. The ten variable select buttons allow the highest level of selection of operator commands.

Referring again to FIG. 14, it can be seen that the logic for processing this menu is contained herein.

Element 686 determines whether a menu change is to be made. If it determines yes, procedure ERASEBOARDS is called at element 688 which clears the display. Element 690 calls procedure TWOSQUARES which draws the two rightmost buttons. Element 692 calls procedure TWELVESQUARES which draws the remaining ten buttons and procedure RIGHTBUTTONS is called at element 694 which labels the two rightmost buttons.

Element 696 calls procedure ERASE which clears menu description 520 from display. Element 698 labels the ten remaining buttons, that is the ten variable buttons 526 in the manner shown in the first entry of Table A. Element 700 calls procedure WAITFORTOUCH. This procedure retains control of general purpose processor 112 until such time as an entire command is received. This command is of course input via program SAMPLE$lNTERRUPT 648 (see also FIG. 13).

Upon receipt and validation of a command, control is returned to element 702 for coding. As can be seen, all ten variable select buttons are defines on the menu. Therefore, there is an element corresponding to each of these ten buttons as shown. For example, element 704 corresponds to button 1, element 708 corresponds to button 2, and so forth for all ten buttons. Reference to the first entry of Table A shows that select button 1 is defined as ECG SETUP. Referring back to FIG. 14, element 706 calls procedure ECGSERVICE in response to button 1 and control is transferred to that procedure which, in turn, presents the ECG MENU. Similarly, depressing button 2 causes element 710 to call procedure DENSSERVICE. Reference to Table A again shows that select button 2 is defined as EAR DENS SETUP.

Button 3 causes element 714 to call procedure PACFTNSERVICE which handles the pacer function menu. Pressing button 4 results in a call to procedure ASSESSRUNSERVICE by element 718. This is the main processing menu and associated programming which performs the mathematics associated with assessing a run.

Depressing button 5 results in a call to procedure ONINSTO by element 722 which sets signal AUTOINSTO to ECG processing 120. A delay is caused by element 724 followed by a call to procedure OFFINSTO by element 726 which clears signal AUTOINSTO. Depressing button 5 therefore is intended to enable a special selection of lead configuration within ECG processing 120 and this selection can be used for maintenance purposes or for other manual intervention.

Depressing button 6 results in a call to procedure OUTPUTSERVICE by element 730. This causes a printing on printer 144 of the information on display 28. Depressing button 7 provides a call by element 734 to procedure OPTION$SERVICE which permits a call to an optional menu. Depressing button 8 is an operator request for printing the contents of the screen of display 28. This is caused by a call to procedure PRINT$START by element 738. Button 9 causes a call to procedure DATASETSERVICE by element 742. This results in the output of a standard buffer to the display. Depressing button 10 is a reques by the operator to advance paper on printer 144. This is accomplished by a call to PRINT ADJUST by element 746 followed by a delay caused by element 748 followed by a second call to PRlNTADJUST by element 750.The first call to procedure PRINT$ADJUST simulates the pressing of the advance paper button, element 748 produces a delay which ip equivalent to that which would be found by manual operation and the second call to procedure PRINT ADJUST simulates the manual release of the advance paper button.

It will be noticed on FIG. 14 that there is no normal access of procedure MENU 684. At the pressing of a button, another menu is called, an operation is performed, and the program returns to element 700 to await a further command. The control therefore remains with procedure MENU 684 unless it is terminated by depressing one of the two right hand buttons resulting in a corresponding change in the INTERRUPT PROCESSING program or may be the result of an abnormal exit from one of the secondary menu processing routines.

FIG. 15 is a schematic for procedure WAITFORTOUCH 752. Notice that it is a very simple routine.

It checks the input buffer prepared by procedure SAMPLE$lNTERRUPT 648 at element 682. See also FIG. 13. Referring again to FIG.15, the buffer is checked at element 754. An input requires all four bytes to be present. Until all four bytes are present, control remains within element 754. The actual input is accomplished in the INTERRUPT routine which scans the output of USART 104 every four msec as explained above. After receipt of all four bytes, element 756 determines whether or not the command is valid. This determination is made based upon the number of buttons defined for a specific menu. In the general menu, wherein all ten variable buttons are defined, any one of the ten inputs is valid.However, as can be seen from Table A, other menus define a lesser number of buttons and validity for that menu requires that a button is depressed which corresponds to the button defined for that menu. After element 756 determines that the input is valid, the exit is made at element 758 and control is returned to the calling program.

FIG. 1 6 shows procedure ECGSERVICE 760 which is called by the depressing of Select 1 for the MAIN MENU. Reference to Table A shows that the ECG MENU is entry number 2. Entry number 2 describes the purpose of the menu and describes the.legend for each of the select buttons. Referring again to FIG.16, it can be seen that element 762 calls procedure ERASE which erases the description of the ten select buttons 526. Element 764 writes the labels associated with the ECG adjustment menu as these are seen on entry 2 of Table A.

Referring again to FIG. 16, it can be seen that element 766 awaits a command by calling procedure WAITFORTOUCH. As with the MAIN MENU, the ECGSERVICE MENU has all ten buttons defined. Therefore, there is an action associated with depressing any of the ten buttons. Button 1 causes selection of the l, ll, III ECG lead configuration. This is accomplished by a call to procedure ONINSTO via element 770 which disables the ECG input AMP 120 to prevent lead change from saturating the amplifier. A call is made to procedure ADJUST by element 772 which sends the select code via l/O 108 to ECG control 118. A call is made to OFFINSTO to resume ECG processing. A call is made to UN$ROTATE at element 774 which records that the data on the physical input cable corresponds to the selected lead configuration.Element 776 merely legends the analog traces. Reference to FIG. 9 shows the legends as elements 510, 512 and 514.

Depressing button 2 requests configuration AVR-AVL-AVF. This is accomplished by a call to procedure ONINSTO by element 780, a call to procedure ADJUST by element 782, a call to OFFINSTO, a call to procedure US NOTATE by element 784, and again writing the proper legends on display 28 at element 786.

Depressing button select 3 causes a similar selection to lead configuration V1 -V2-V3.

Similarly, depressing button 4 causes selection of lead V4-V5-V6 and depressing button 5 enables the selection of the special lead configuration. Notice that these configuration procedures are all quite similar and are accomplished in a very similar fashion.

Depressing button 6 causes a clearing of the digital portion of ECG processing 120. This is accomplished by a call to procedure ONINSTO at element 826, a delay at element 828 and a call to procedure OFFINSTO at element 830. Pressing button 7 allows the operator to rotate the analog ECG traces as displayed on the screen of display 28. This assists in comparing these traces to the analog traces from the densitometer input. This rotation is accomplished by keeping track of the present position within a variable called ROTATE I. Element 834 decides whether or not that variable is equal to 1. If it is, configuration 2 is called at element 836 and the labels on display 28 are changed by element 838 calling procedure ROTATE$TEXT. Should element 834 find that variable ROTATE I does not equal 1, element 840 determines whether ROTATE I is equal to 2. If it is, element 842 changes to configuration 3 and element 844 calls procedure ROTATE$TEXT to change the labels. Otherwise, element 846 selects configuration 1.

Buttons 8 and 9 are used to increase and decrease the gain of the ECG channels. Button 8, if depressed, requests an increase in ECG gain. Element 852 determines whether the maximum gain has been reached. If so, the operator is notified by element 858 that the gain is presently at maximum. If the gain is not at a maximum, element 854 increments the gain value and element 856 calls procedure CHANGEGAIN to output the change in gain. As explained earlier, with reference to FIG. 3, the gain change is communicated from l/O 108 to level control 134 via cable 128 as a two-bit quantity and to ECG control 11 8 as an additional two-bit quantity. Level control 134 utilizes this two-bit quantity to control the gain of each of three ECG channels. Depressing button 9 similarly causes a decrease in ECG gain.Element 862 determines whether the minimum gain has been reached. If it has, the operator is notified via a signal at element 868. If the minimum gain has not been reached, element 864 decrements the gain and element 866 calls procedure CHANGEGAIN to output the gain change to level control 134 and ECG control 118.

The button 10 is used to return control to the program after ECG adjustments have been made.

The element 872 determines whether or not a run is in progress. If the run is not in progress, control is transferred to procedure MENU via element 876 which corresponds to a normal transfer of control. If the run is in progress, control is transferred via element 874 to procedure HALT$MENU which corresponds to a termination of an existing run and facilitates preparation of the menu. As can be seen from FIG.16, there is no normal exit from ECGSERVICE 760. Control is either retained by this procedure or it is returned by procedure calls to MENU or HALT$MENU.

FIG. 17 is a flowchart for procedure DENSSERVICE 878. This procedure is located within module ECGDEN as shown in Table B. As shown in Table A, this procedure is called by depressing select 2 when in the MAIN MENU. The function of the procedure is to perform those adjustments for the densitometer sensor in an analogous way to those performed for the ECG.

Element 880 calls procedure ERASE which clears the ten select buttons. Element 882 writes the labels on the first five select buttons required by this menu. See also Table A. Element 884 calls procedure WAITFORTOUCH which awaits a valid command. In this case, the valid command corresponds to a depressing of one of the first five buttons, that is select 1 through select 5. The depressing of button 1 requests an increase in the gain of the densitometer analog channel. Element 888 determines whether the maximum densitometer gain has been reached. If not, the gain of that channel is incremented by procedure DEN$ADJUST as called by element 890. Similarly, depressing button 2 causes the decrease in the gain of the analog densitometer channel. Element 94 determines whether the minimum gain has been reached.If not, element 896 calls procedure DEN ADJUST which decrements the gain of the channel.

The gain of the derivative of the densitometer signal is increased as a result of select number 3, elements 900 and 902 and procedure D$DEN$ADJUST as shown. In a similar fashion, the gain of the derivative channel is decreased by select number 4 and elements 906 and 908. Select number 5 allows termination of the run and the pressing of select number 5 results in element 912 determining whether a run is in process. If not, normal control is transferred to procedure MENU by a call from element 914. If a run is in progress, however, element 916 calls procedure HALT$MENU which makes the call for the abnormal exit from the procedure.

FIG. 18 shows that procedure PACFTNSERVICE 91 8 is not yet written. A call is made to procedure NOTWRITTEN via element 920 which notifies the operator of that fact.

The majority of the processing done by the NIPPA system is done as a result of a procedure call to procedure ASSESSRUNSERVICE 924. By reference to Table A, one can see that the procedure is called from the MAIN MENU by select number 4. Reference to Table B shows that the procedure is located within module MENRUN.

Referring to FIG. 1 9, element 926 determines whether a run is currently in progress. If the answer is yes, element 928 calls procedure RESTART. As is shown in FIG.19, this procedure is not currently available. Element 930 calls procedure ERASE which clears the ten select buttons. Element 932 writes the new legends on the buttons and also provides the menu description. Reference to Table A shows that entry number 5 is produced at this time. Notice that the menu description is SELECT NUMBER OF RR INTERVALS DESIRED. The buttons are legended 2, 5, 10, 20, 50, 100, 180 and 240 respectively.

Notice that only eight buttons are required. Control is then transferred via element 934 to procedure WAITFORTOUCH which awaits the operator selection.

Following operator selection, element 936 calls procedure ERASE which erases the ten select buttons and any description. Element 938 writes the new menu description and the new legends on the buttons. According to Table A, one can see that entry number 6 of Table A is this new menu. The menu description becomes SELECT CHANNEL FOR PWAVE DETECTION and that select 1, 2 and 3 buttons become top channel, second channel and composite respectively. The remaining seven buttons are not used. Control is then again transferred to procedure WAITFORTOUCH by element 940.

After valid operator selection of the desired options, element 942 clears the select buttons and the menu description and element 944 writes the new menu description and select 1 button is shown in entry 7 of Table A. Notice that this menu merely allows the starting of the assessment process. Control is transferred to procedure WAITFORTOUCH by element 946 and the start command is awaited.

Following the start command, element 948 again calls procedure ERASE, which clears the button and menu description. Element 950 writes the next menu which is shown as entry 8 of Table A. The menu description becomes RUN IN PROGRESS NUMBER OF INTERVALS DETECTED = This allows the program to update the number of intervals detected as it progresses. The single select button, that is select 1 which is used, allows the operator to stop the run.

Element 952 calls procedure REFERENCE$AMPLITUDE8OF$ECG. The purpose of this procedure is to establish a reference amplitude for the ECG signal. Element 954 writes the number of cycles into the blank provided in the menu description. Element 956 performs a search of the buffer and element 958 calls procedure DATA$ACQUISITION. This procedure synchronously moves the analog data on display 28 as it is searched. Element 960 determines whether or not analog data is to be printed. If it is to be printed, element 962 starts the printer 144 by calling procedure NO$D4iPRINT$START.

Element 964 again calls procedure REFERENCE$AMPLITUDE$0F$ECG and element 966 calls procedure INITIAL IZE$RWAVESENSING. This procedure initializes the search parameters for R-wave detection. Procedure CALIBRATE is a call by element 968 which establishes a pulse rate window. A delay is then achieved by element 970.

The element 972 calls procedure ERASE which clears the stop button. Element 974 displays status to the operator. Element 976 calls procedure SAM$START which begins processing of the ECG signal. Procedure ECG+DEN$MEA$;UREMENT is called by element 978 which correlates the ECG and densitometer inputs.

The major calculations are performed by procedure PARAMETERS$CALCULATlON as called by element 980. These data as calculated are then evaluated by procedure PARAMETERS$EVALUATION as called by element 982.

As this signifies the end of the calculations, a new menu is prepared. Element 984 calls procedure ERASE which clears the buttons in the menu description and element 986 notifies the operator that the run has been completed. Item 9 of Table A shows the menu that is produced. Notice that the menu description is RUN COMPLETED SELECT DISPLAY PROGRAM. Qnly three select buttons are required, giving the operator the option of displaying tabular data, graphical data or raw data. Element 990 calls procedure WAITFORTOUCH and awaits the operator selection of the output option. Elements 992, 996 and 1000 decode the operator selection of output format. If select 1 was chosen, the data will be displayed in a tabular format. Element 994 calls procedure TABLEDISPLAY. FIG. 27 shows the data as displayed in a tabular format. This format is discussed in greater detail below.

Referring back to FIG.19, selection 2 provides for graphic display of the run data. The graphic display is created by element 998 calling procedure GRAPHICDISPLAY. The third selection is for a raw data display which is accomplished by element 1002. with a call to procedure RAWtDATAtDlSPLAY.

Exit is made at element 1004.

FIG. 20 shows a detailed flowchart of procedure HALT$MENU 1006. As mentioned above, this procedure is used for the abnormal termination of an assessment run or other processing operations.

Element 1008 determines whether or not any buttons have been drawn. This determination is made by a simple flag, which is stored withinmemory 106. If it is determined that no buttons have been drawn, element 1010 calls procedure ERASEBOARDS which completely clears the display. Element 1012 then calls procedure TWOSQUARES which draws the two rightmost system level buttons. Element 1014 calls procedure TWELVESQUARES which draws the remaining ten variable select buttons. Procedure RIGHTBUTTONS is called by element 101 6 which labels the two rightmost buttons.

Element 1018 determines whether or not the run in progress was completed. If it was not, element 1020 calls ERASE to clear the buttons and element 1022 writes a menu on display 28 which notifies the operator that the run was interrupted. Reference to Table A, item 10 shows the menu corresponding to an interrupted run. The menu description becomes RUN INTERRUPTED SELECT ACTION DESIRED. Element 1024 labels the select buttons which are similar to the buttons used for the MAIN MENU, allowing maximum flexibility of operator options.

If element 1018 determines that the run was complete, element 1026 prepares the RUN COMPLETE menu which is item 9 of Table A notifying the operator that the run in progress was completed. Notice that the buttons, however, will remain the same which permits maximum operator selection. Element 1028 transfers control to procedure WAITFORTOUCH awaiting the operator input command. Again, all ten select buttons are defined for this menu which means that there are ten separate possible branches within the program. If select 1 is chosen, element 1032 determines whether a run was interrupted. If the run was interrupted, element 1 034 clears the buttons and element 1 036 writes WAIT. Element 1038 calls procedure PARAMETERS$CALCULATION which calculates the data from the previous run's input.Element 1040 calls procedure PARAMETERSEVALUATlON which then evaluates the data. Element 1042 calls TABLEDtSPLAY which displays the data in tabular format.

Select 1 corresponds to button display results shown in item 10, select 1 in Table A.

Selection of button 2 causes element 1046 to call procedure ECGSERVICE which produces an ECG menu. Selection of button 3 similarly causes element 1050 to call procedure DENSSERVICE which produces the densitometer menu and permits adjustn ents to the densitometer channels. Selecting button 4 causes element 1054 to call procedure RUN RESTART which is the restart menu. As explained above, this function is not currently available. Select 5 causes element 1058 to call procedure ASSESSRUNSERVICE which performs the calculation on the run data. Selection 6 causes a return to procedure MENU which is a return to the MAIN menu.

Select 7 is a request to display raw data. Element 1066 calls procedure RAW$DATA$DISPLAY.

Select 8 is a print request and it causes procedure PRINT$START to be called bv element 1070. Element 1074 calls procedure DATAtSET$SERVICE in response to select 9. This display is the standard data set. Select 10 is an INSTO request. This is honored by element 1078 calling procedure ONINSTO.

Element 1080 delays corresponding to the normal manual operator input. Element 1082 calls procedure OFFINSTO. Thus it can be seen that procedure HALT$MENU 1006 is a menu quite similar to the MAIN menu except that it allows for recovery from an interrupted program run and therefore assumes that there exists some input data already in the buffer. 1084. As can be buffer FIG. 21 is a flowchart for procedure REFERENCE4;AMPLITUDE;OFi;ECG 1084. As can be seen in Table B, this procedure is located within module PQRDD. Element 1086 initializes certain variables.

Element 1088 determines whether all the data is scanned. If it has been scanned, element 1090 outputs the reference and element 1092 provides an exit to the calling routine.

If element 1088 determines that not all data has been scanned, element 1094 increments the pointer and element 1096 determines whether all the densitometer points have been scanned. If they have, control is returned to element 1088. If all densitometer points have not been scanned, then element 1098 determines whether the present value is larger than the temporary established reference.

If it has not, control is returned to element 1096 to be determined whether all densitometer points have been scanned. If element 1098 determines that the value is larger, element 1100 establishes a new temporary reference and returns control back to element 1096. It is the function of this procedure to establish a running reference.

FIG. 22 is a flowchart of procedure SAM;,tSTART 1102. As can be seen from Table B, this procedure is located within module PQRDD. Element 1104 initializes certain variables. Element 1106 calls procedure FIND$NEXT$R. This procedure locates the very next R wave. Element 1108 determines whether the buffer has been completely searched. If the answer is yes, element 1110 calls procedure DATA$ACQUISITiON which unfreezes the analog traces on the display screen and loads the next 8.5 seconds of data. Element 1112 calls procedure SAM$START which causes recursive entry of the present procedure.

Element 1114 determines whether the R-wave location is valid. If the answer is no, element 1106 calls procedure FIND$NEXT$R which locates the next R wave. If the R wave location is valid, element 111 8 records the location. Element 1120 again calls procedure FIND$NEXT$R which locates the next R wave.

Element 11 22 determines whether the buffer has been completely searched. If the answer is yes, element 1110 calls procedure DATAACQUlSITlON which again unfreezes all the analog traces and 8.5 additional seconds are loaded into the buffers and display. Procedure SAM$START is again called by element 1112. Exit from the procedure is via element 1128.

FIG. 23 provides a detailed flowchart of procedure ECG$DEN$MEASUREMENT 1130. This procedure correlates the ECG and densitometer input. The procedure is located within module PQRDD.

Element 11 32 fetches the next R wave as previously found by procedure SAMSSTART 1102 (see also FIG. 22). Element 11 34 calls procedure SAM$STOP which finds the following R to ensure that full cycle exists in the buffer. If no following R exists, SAM$STOP refills the buffer and continues the processing.

Procedure RgPEAK$LOCATION is then called by element 1136. This procedure defines the next R peak.

Element 11 38 next calls procedure P$PEAK$LOCATION which determines the location of the corresponding P wave. Procedure D4iDEN$MINtIiMAX is called by element 1140 which finds the minimums and maximums of the derivative of the densitometer for the cycle of the interest.

Element 1142 calls procedure D$DEN$MIN$LOCATION which finds the location of the minimum densitometer derivative within the cycle of interest. Procedure SPlKEDATA is called by element 1144 which finds the atrial pacing artifact. Procedure SPIKE$DATA is again called by element 1146 to find the ventricular pacing artifact. Element 1148 determines whether all cycles have been processed by determining whether all R waves have been found. If not, the processing is continued. After all R waves have been found, exit is made via element 11 50.

FIG. 24 shows the logic within SAMSTOP 11 52. Element 11 54 calls procedure FIND$NEXT$R which locates the next R wave. Element 1 56 determines whether all R waves have been found. If the end has not been determined, element 11 64 outputs the number detected which is written onto display 28 in the blank space within the menu description, notifying the operator of the progress. As may be seen from entry 8 of Table A, the menu description within the description describes RUN IN PROGRESS NUMBER OF INTERVALS DETECTED = -------. .Exit is made via element 1166.

If element 11 56 has determined that the end has been reached, element 11 58 calls procedure DATA$ACQUISITION which unfreezes the current analog trace and loads the next 8.5 seconds of information. Element 11 60 calls procedure SAM$START which begins processing the ECG information.

Element 11 62 again calls procedure SAM$STOP to continue the processing of data.

The major calculations for run assessment are performed by procedure PARAMETERSS CALCULATION 11 68. A flowchart of the procedure is found in FIG. 25. This procedure is located within module MATH. Elment 1170 initializes various parameters. There are several internal loops which are to be used within a procedure. The element 1172 determines whether or not the Nth P wave is within range. If it is not, element 11 76 records a zero for the Nth P-Q and P-R intervals. If the Nth P wave is found to be within range, element 11 74 records the P-Q and P-R intervals determined. Element 11 78 determines whether the first cycle is being processed.If the answer is yes, element 11 80 computes the value of R to R only, and control is transferred to element 11 94.

For every one but the first cycle, element 11 82 determines and records the RR interval, the preejection phase (PEP), the left ventricular ejection time (LVET), and the rate corrected LVET. To accomplish this, it determines that the Nth PEP is equal to the time at which the densitometer amplitude is a minimum minus the Q wave for that same cycle. It also determines that LVET is equal to the point at which the derivative of the densitometer is at a minimum minus the densitometer minimum.

Element 11 84 determines whether the RR value is too large. Notice that this would correspond to a very slow heart rate. If the RR interval is too large, a correction factor is applied by element 11 86 which represents the maximum correction factor. If the RR interval is within range, however, element 11 88 uses a correction factor determined from an internal table. These correction factors are necessary because of the time delay between the electrically sensed ECG signals from the region of the heart and the time it takes for the densitometer at the ear to register the corresponding arterial pulse.

Element 11 90 determines the corrected Nth LVET by dividing it by the selected correction factor.

Element 11 92 determines the densitometer amplitude as being the maximum densitometer amplitude for that cycle minus the minimum densitometer for the same cycle. Element 11 94 determines whether all N cycles have been completed. If not, control is returned to element 11 72. After all cycles have been completed, exit is provided via element 11 96.

FIG. 26 is a detailed flowchart for procedure PARAMETERSEVALUATlON 1198 which is a procedure within module MATH as shown by Table B. This procedure compiles the tabulated statistics for the raw data measured by the previously described procedure. Element 1200 sets index K=(i.

Element 1202 establishes an interim value equal to the minimum PQ divided by the sampling interval plus the PQ increment divided by the sampling interval times the value of K. This is done by the Kth cycle. Element 1204 determines whether all PQs have been evaluated. Element 1206 clears index M.

Element 1208 clears index N.

Element 1210 determines whether the Nth PQ interval is less than Mth PQ interval. If it is, element 1212 adds the Nth densitometer amplitude to the running sum of densitometer amplitudes for the Mth cycle. Element 1214 adds the Nth PEP to the Mth running sum of PEP. Element 1216 adds the Nth LVET to the Mth sum of LVET. Similarly, element 121 8 adds the Mth PEP LVET ratio to the Mth sum of PEP LVET ratio.

Element 1220 determines whether all N cycles have been reviewed. If so, element 1222 determines whether all six Ms have been tried. If not, control is returned to element 1208. After all M have been tried, element 1224 determines the average Nth sum of the densitometer amplitudes, the Nth sum of the PEP's, the Nth sum of the LVET's, and the Nth sum of the PEP LVET ratios. Element 1226 determines if all the N averages have been computed. If not, control is returned to element 1224.

After all have been computed, element 1228 normalizes all N PEP's, densitometer amplitudes and the LVET's. Exit is performed at element 1230.

FIG. 27 shows the format for the tabular data display. This format shows the cycle number as beat number on the left hand side. The columns to the right show RR interval in milliseconds, PQ interval in milliseconds, PR interval in milliseconds, PEP in milliseconds, PEP LVET ratio, and the detection of atrial and ventricular pacing artifacts.

FIG. 28 shows the format for the raw data display. Again, the cycle number is shown in the left most column. The raw data displayed are intermediate computational products. These values correspond directly to values produced within various NIPPA system programs. From left to right, they include the R to R reference level. This allows the NIPPA program to establish a variable window for the determination of R to R interval. The R wave, Q wave and P wave substitutes are next shown. This is followed by column for the maximum densitometer amplitude. The densitometer minimum substitute is shown followed by the densitometer minimum and the densitometer derivative minimum substitute.

These raw data values have application primarily for maintenance and test purposes. It can be seen by those of ordinary skill in the art that the present invention just described as embodied in the NIPPA system has application to many other monitoring and assessment systems. Such applications are readily apparent from the system as just described.

Claims (13)

1. Apparatus for determining a physiological parameter of a patient comprising: first means for generating a first signal indicative of quantity of blood in a portion of said patient; second means for generating a second signal indicative of electrical activity of an organ of said patient; and means responsively coupled to said first generating means and said second generating means for computing said physiological parameter using said first signal and said second signal.

2. Apparatus as claimed in claim 1 wherein said computing means further comprises: means responsively coupled to said first generating means for differentiating said first signal.

3 Apparatus as claimed in claim 1 or 2 wherein said second signal is an electrocardiogram.

4. Apparatus as claimed in claim 1,2 or 3 wherein said first generating means is a densitometer.

5. Apparatus as claimed in claim 4 wherein said densitometer is an ear densitometer.

6. Apparatus as claimed in claims 2, 3 and 4 wherein said computing means further comprises: first means responsively coupled to said densitometer for processing said first signal; second means responsively coupled to said differentiating means for processing a derivative of said first signal; third means responsively coupled to said second generating means for processing said electrocardiogram; an analog-to-digital converter responsively coupled to said first processing means, said second processing means and said third processing means whereby said first signal, said derivative of said first signal, and said electrocardiogram are converted to digital representations; and a processor responsively coupled to said analog-to-digital converter whereby said digital representations are converted to said physiological parameter.

7. Apparatus as claimed in claim 6 wherein said computing means further comprises: means responsively coupled to said processing means for displaying said physiological parameter.

8. Apparatus as claimed in any preceding claim wherein said physiological parameter includes pre-ejection phase.

9. Apparatus as claimed in any preceding claim wherein said physiological parameter includes left ventricular ejection time.

10. Apparatus as claimed in claims 8 and 9 wherein said physiological parameter further includes the ration of pre-ejection phase and left ventricular ejection time.

11. Apparatus as claimed in claim 10 further comprising: means responsively coupled to said third processing means for identifying a pacing pulse whereby hemodynamic effects of pacing may be determined.

12. Apparatus as claimed in claim 6 or 7 further comprising: means responsively coupled to said third processing means for identifying a pacing pulse.

13. Apparatus for determining a physiological parameter of a patient, substantially as hereinbefore described with reference to the accompanying drawings.