The National Aeronautics and Space Administration variety of programs covering many different disciplines and many different objectives. In this brief overview, all of the programs of NASA cannot be covered, but an attempt will be made to describe some of the diverse activities within the agency. Toward the end of this paper, attention will be directed to those activities which are related to the medical profession. Detailed discussion of work closely related to cardiovascular problems will be avoided since these problems receive the primary emphasis of this book. Other papers will discuss in considerable detail NASA programs related to the cardiovascular area.

Over the past 45 years angiocardiography has become established as the most accurate method for clinically defining the size and shape of vascular and cardiac structures. Throughout this period it has received widespread use for the diagnosis and treatment of peripheral vascular disease, congenital and valvular heart disease, and most recently, in dealing with coronary artery lesions. These uses continue despite recent technologic advances in other less invasive diagnostic techniques, such as echocardiography and nuclear medicine. In fact, angio-cardiography will continue to serve, as it has served in the past, as the principal standard of reference in man for calibration and/or comparison of newer methods for determining cardiac volume or dimensional change.

Conventional angiocardiography is classified as an "invasive" method, since the contrast material is injected selectively via catheters into the central circulation. This provides maximum contrast with minimum superposition. The resulting angiocardiograms represent the single-, bi- or multiplane x-ray projection of the instantaneous spatial dye distribution, frame by frame. The new technique of computerized video-angiocardiography to be described minimizes the disadvantages of contrast material and provides the maximum structural and/or functional information from angiocardiograms for a given amount of contrast independent of the site and mode of injection. Even with 1/5 to 1/10 of the normally used amount of contrast material and peripheral injection, the con-tracting heart and the vessels can be clearly visualized. This improvement is reached by summing up instantaneous contrast angiocardiographic pictures in a continuous or heart phase related mode, and subtracting the corresponding non-specific background obtained from x-ray pictures taken in real time prior to or following contrast injection. Maximum flexibility of the "gated" integration and subtraction angiocardiography is facilitated by digitizing complete videoangiocardiograms, each field in real time, and by storing it into the mass memory of a digital computer for further image processing. Digital filtering and histogram modification techniques are used to enhance the contrast, and to reduce the "noise." Our technique as presently used and some typical results will be demonstrated. If this computerized angiocardiography is performed in its "invasive" version, higher contrast and improved border recognition can be achieved with even less dye and circulatory disturbances. Quantitative videoangiocardiographic techniques for volume, flow, and contraction pattern analysis can be facilitated by the described modes of digital image preprocessing.

Restoration of coronary blood flow following successful bypass surgery would be expected to lead to improvement of regional contraction. Several methods have been described for the quantitative assessment of regional or segmental left ventricular wall motion from the left ventriculogram. All methods need a reference point as a basis for their coordinate systems or to compensate for movement of the patient relative to the film camera. Morphological abnormalities such as aneurysms or akinetic areas influence the location of the reference point, resulting in artifacts when measuring wall motion. Localized abnormalities of contraction are not always reflected in "overall" measurements. Akinesis and/or dyskinesis of one or more segments can occur in the presence of normal end diastolic volume or ejection fraction. Such overall parameters will, however, gradually go into pathological ranges with an increase in the number of abnormally contracting segments. Another approach to the quantification of local wall motion uses radiopaque markers. In this study marker pairs are implanted during bypass surgery in the area of newly perfused regions as well as in control regions. Sequential cineradiograms at 50 fr. per second were made at intervals over a period of one year. It was found that direct traumatic effects of the surgical intervention overwhelm the expected improvement of myocardial function in the first three postoperative months. At present the quantitative approach to segmental cardiac function is mainly one of image analysis in one or two preselected planes. Subjective visual interpretation of these images should be replaced by objective data analysis.

A computer technique is being developed at the Jet Propulsion Laboratory to automate the measurement of coronary stenosis. A Vanguard 35mm film transport is optically coupled to a Spatial Data System vidicon/digitizer which in turn is con-trolled by a DEC PDP 11/55 computer. Programs have been developed to track the edges of the arterial shadow, to locate normal and atherosclerotic vessel sections and to measure percent stenosis. Multiple frame analysis techniques are being investigated that involve on the one hand, averaging stenosis measurements from adjacent frames, and on the other hand, averaging adjacent frame images directly and then measuring stenosis from the averaged image. For the latter case, geometric transformations are used to force registration of vessel images whose spatial ori-entation changes.

Both nuclear and computerized cardiovascular imaging are being evaluated for use in assessing myocardial integrity. The general approach used in assessing myocardial integrity by positron emission tomography consists of the administration of a radiopharmaceutical label with a positron-emitting radionuclide followed by the imaging of the distribution of the label within the myocardium as a func-tion of time. Spatial resolution is currently limited to approximately 1 cm and the measurements currently require several minutes. Although technological advances in transmission computed tomography have been dramatic, its myocardial applications are currently limited by inadequate temporal resolution and difficul-ties in timing the scan exposure to coincide with maximum contrast enhancement. Dynamic scanning (multiple scans in rapid succession), gated, and rapid scans (less than 500 milliseconds) are currently being developed. Modest improvements can be expected with existing technology and optimization of reconstruction algorithms.

Positron emission tomography (PET) is a technique which yields transverse tomographic images of the body reflecting the distribution of a positron-emitting radionuclide. When used in conjunction with certain radionuclides incorporated into molecules of physiological importance, PET is a promising method for the assessment of myocardial integrity through the mapping of biochemical processes.

Although the technological advances in transmission computed tomography (CT) have been dramatic, cardiovascular applications of CT are currently limited by inadequate temporal resolution to resolve moving structures (e.g., the myocardium), difficulties in timing the scan exposure to coincide with maximum contrast en-hancement of vascular structures and cardiac chambers, and limited field of view perpendicular to the axial plane. The modes of cardiac CT operation are reviewed: static (single scan), dynamic (multiple scans in rapid succession), gated and rapid (scans <500 msec). Modest improvements in gated and dynamic scanning are expected with existing technology and optimization of reconstruction algorithms. Future advances in CT technology should be directed toward the development of rapid scanning systems for imaging the myocardium and toward higher resolution scanners for imaging vascular structures.

Film subtraction angiography from intravenous injections of iodinated contrast media was one of the earliest radiographic methods used to visualize the heart and blood vessels. Initially used for studies to visualize the cardiac chambers and the aorta, this technique has been supplanted by the more sophisticated (and invasive) methods of selective catheter angiography because of the superior images obtained by virtue of higher intravascular iodine concentrations.

A system is under development for a relatively non-invasive technique for the assessment of atherosclerosis. The principle of this method is digital video x-ray subtraction for the visualization of arterial structures after the intravenous injection of contrast media. The prototype unit for the development of video subtraction techniques has been assembled and preliminary testing has started. Re-sults so far in dogs have shown good visualization of the heart, carotid arteries and renal arteries.

Existing radiographic imaging systems provide images which represent an integration or averaging over the energy spectrum. In order to provide noninvasive angiography it is necessary to image the relatively small amounts of iodine which are available following an intravenous administration. This is accomplished by making use of the special spectral characteristics of iodine. Two methods will be presented. One involves a special grating for encoding the iodine information in the form of a fine line pattern. This is subsequently decoded to provide images of iodinated structures which are otherwise almost invisible. The second method utilizes a scanned x-ray beam which is rapidly switched in the high energy region. In this region, iodine experiences significant variations in the attenuation coefficient while bone and soft tissue do not. An efficient and accurate x-ray detector can be used with scanned x-ray beams. This provides a high degree of sensitivity enabling the visualization of small vessels containing relatively dilute iodine.

A computerized fluoroscopy system employing several time and energy subtraction algorithms has permitted good visualization of the cardiovascular system using peripheral intravenous iodine injections of about 1 cm3/kg. Image contrast im-provements of 8-16 over conventional fluoroscopy are common. Several canine and human imaging studies are described including visualization of myocardial infarc-tions as regions of anomalous image grey shade. The system employs a standard image intensified fluoroscopy system and a specially constructed real-time image processor. Quasimonoenergetic x-ray beams formed by filtration deliver typical doses of 400 mR/sec in adult human cardiac exams.

Modern technology has effected major improvements in health care delivery, yet basic concepts for noninvasive blood pressure measurement have not changed significantly in 60 years. Sphygmomanometry has been improved by electronic pulse detectors, ultrasonic wall-motion sensors, automatic cycling, digital displays, etc.; however, little progress has been made to permit continuous, noninvasive ambulatory monitoring. Increased awareness of hypertension by the medical community and the general public has augmented the demand for reliable, easily administered methods. The three papers in this section demonstrate novel approaches toward improved blood pressure measurement.

Some results of a study of techniques for the noninvasive and ambulatory measurement of blood pressure are presented. A method for computer processing ambulatory or stress test data to determine blood pressure is described. It appears that good accuracies can be obtained with ambulatory patients during normal ranges of physical activities and perhaps during treadmill tests. The development and testing of a new transducer for the noninvasive beat-by-beat measurement of blood pressure is described. The transducer outputs are similar to intraarterial wave-forms that are obtained by catheter, and it is expected that the transducer will be used in the operating room, the ICU, and the CCU.

Analyses are developed for new procedures of (I) noninvasive discrete measurements of the brachial arterial blood pressure by phono cuff-sphygmomanometry (II) noninvasive continuous measurement of arterial blood pressure by arterial ultrasonic imaging. For the phono cuff-sphgmomanometry technique, analyses are developed for the vibration of the fluid-filled cuff-compressed arterial tube and for pressure pulse wave propagation through it, to derive an expression for the corrected brachial arterial pressure in terms of the monitorable auscultatory cuff pressure, the cuff width, the arm size and the principal frequency of the Korotkoff sounds. For the ultrasonic imaging technique, a large defor-mation analysis of the pressurized fluid-filled cylindrical nonlinearly elastic arterial tube is performed, to derive expressions for the arterial pressure and then for the pulse wave velocity (by invoking the distensibility definition) in terms of the arterial tube's nonlinear elasticity parameters (of the strain energy density function), and the unpressurized and deformed tube radii. Monitoring of the arterial radii and the associated pulse velocities at three dif-ferent instants yields, from the pulse wave velocity expression, adequate equations to determine the values of the unpressurized arterial tube radii and its elasticity parameters; they are substituted in the arterial pressure expression, to yield the arterial pressure continuously in terms of the continuously ultra-sonically monitorable arterial diameter.

In a recent paper a new technique for a noninvasive measurement of cardiac pressure utilizing tiny air bubbles was proposed and researched. In the present paper we report on the amplitude and frequency sensitivity to pressure of two types of spheres. Further, we demonstrate the viability of the proposed scheme as applied to remote cardiac pressure measurements.

The ideal diagnostic instrument provides definitive data on the condition of a patient, causes him no harm or discomfort, and must be convenient, reliable, and economical for a physician or his paramedical associates to operate. Such an in-strument must, of course, be totally noninvasive or transcutaneous. Ultrasonic imaging systems exhibit all of the basic characteristics of an ideal diagnostic instrument and unmistakably offer enormous opportunities for further improvements in performance.

Signal processing in diagnostic ultrasound has played a vital part in improving the capability of the instrumentation for imaging. Some current signal processing methods are reviewed and the importance of the transducer in determining the characteristics of the system is emphasized. The possibility of using deconvolution in a single transducer system is discussed and some future developments using digital techniques are outlined.

Standard M-mode echocardiography has been accepted for accuracy and safety proven in over twenty years of clinical use. However the special training needed to interpret echocardiographic tracings in this form has made the technique under-utilized by the general medical public. The extension of echocardiography into two-dimensional ultrasonic imaging permits non-invasive visualization of the heart with an anatomic format that makes the resulting information more comprehensible to the average cardiologist and cardiovascular surgeon. Theoretically, examining signals from all areas of the cardiac valve or chamber reduces the chance of sampling error and in fact provides a more solid base for interpretation of abnormalities. The two-dimensional echo (2DE) instruments provide M-mode information, as well as two-dimensional display, and have been shown to be accurate for quantitation of the size of cardiac valves, the presence of intracardiac masses, the volume of cardiac chambers, segmental left ventricular wall motion abnormalities, complex congenital cardiac malformations, and the presence of pericardial fluid surrounding the heart. Several laboratories are carrying out exciting investigations on the effects of acute and chronic coronary artery disease on the motion pattern of ventricular walls as well as recognizing the main coronary vessels per se. Seeing the main coronary arteries allows attempted recognition of pathology within these vessels directly and serves as a basis for identifying the position of intracardiac and transcutaneous flow measuring devices. Also, 2DE can be used to identify the orientation and location of myocardial segments to be sampled for tissue signature analysis in attempts to non-invasively characterize the histology of the heart.

The papers presented in this section are a small sample of a gradually evolving area of cardiovascular diagnosis. Blood flow detection using Doppler techniques was originally demonstrated in the late 1950s and early 60s; and although overall progress to date seems to be far behind M-mode echo and 2D imaging clinical applications, more rapid progress appears to be looming on the horizon.

Quantitative noninvasive assessment of cardiac physiologic variables in a clinical setting is a desirable yet technically difficult goal. The classical param-eters of interest include chamber or vessel pressures, ventricular dimensions and volumes, and blood flow. Ultrasonic techniques have demonstrated the potential for dimension and flow detection; however, direct noninvasive pressure measure-ments are still beyond the state of the art for this modality. The Ultrasound Program at the University of Washington has-as its major goal the development and application of new ultrasonic technologies to the problem of quantitative cardiovascular assessment. The current status of this program will be reviewed with particular emphasis on cardiac applications. Experience gained from similar pe-ripheral vascular applications will be used to demonstrate the approach. A real-time computer-based ultrasound system with specialized displays is being developed. Components of this system include real-time cross-sectional imaging, pulsed Doppler flow detection and imaging, transducer position locator devices, digital image storage and display units and microprocessor-based signal analysis techniques.

During the last 20 years, Doppler ultrasound has become widely accepted as a means for qualitative or semi-quantitative assessment of blood-flow patterns. Application to transcutaneous quantitation of volume flow (in milliliters/sec), however, has proved difficult. Generally, it has been assumed that the angle between the vessel axis and ultrasound beam must be determined precisely through imaging or triangulation. In addition, because volume flow is a product of velocity and area, the true area of the vessel lumen has been assumed to be essential to volume flow estimation. It has also been considered necessary to measure the velocity profile across the lumen on a point-by-point basis. A new approach to volume-flow measurement is being developed to avoid these problems, wherein flow is measured perpendicular to an arbitrary tomographic plane. The area of the lumen (as projected onto the measurement plane) and the area-average velocity of the fluid per-pendicular to this plane are the only two characteristics relevant to the estimation of volume flow. The product of these two quantities is, by definition, volume flow. As a result, precise measurements of vessel orientation and true diameter and details of the velocity profile are not required.

Although measurement of magnetic fields has played an important role in physics and geology, until recently it has not been an important factor in clinical measurements. The reason for this is that the magnetic field produced by currents flowing in the human body is extremely small compared with the earth's magnetic field and the magnetic noise associated with urban backgrounds. The magnetic fields produced by currents flowing in the human heart are 106 smaller than the earth's magnetic field. With the development of superconducting SQUID magnetometers and superconducting circuitry capable of measuring a gradient of a very small magnetic signal in the presence of a large background magnetic field, it has become practical to record the magnetocardiogram of the human heart.

Recent developments in the design and application of SQUID magnetometers to biomagnetic measurement is reported. It has become well established that magnetic fields are produced by electrical currents associated with muscle and nerve polarization and depolarization. Since magnetic measurements are noninvasive and body tissues are magnetically transparent, biomagnetic techniques are being explored as a promising new observational method. Magnetic fields produced by the human body have very low strength; by compari-son, Earth's ambient field is 106 larger. Definitive measurement of such weak signals in the presence of large, and in most cases unavoidable, ambient noise levels has placed difficult but not insurmountable requirements on the design and use of magnetic instrumentation. Recent developments in SQUID sensor technology have resulted in significantly reduced susceptibility to noise and increased sensitivity. SQUID sensors are currently available that achieve a high level of common-mode noise cancellation. An overview of the basic principles used to achieve this noise reduction, the types of sensors available, and the laboratory techniques for human biomagnetic measurement is presented.

The development of sophisticated data analysis techniques, coupled with dramatic improvements in MCG instrumentation, have made it possible to use the magnetocar-diogram for quantitative, clinical assessment of cardiac electrical function. There are currently four areas where magnetocardiography might prove clinically useful: to complement ECG diagnosis, to screen large numbers of people for cardiac electrical abnormalities, to measure non-invasively the electrical activity of the Bundle of His, and to detect infarct-related injury currents. The success of these applications will depend in part upon obtaining a detailed understanding of the production and detection of the cardiac magnetic field and the relation between the ECG and MCG. Research groups in Finland, France, and the United States are using visual examination of simple MCG signals and complex surface maps, statistical analysis of the morphology of normal and abnormal MCG data, and various equivalent generators to determine how the ECG and MCG are related and to identify those areas where magnetocardiography has the most promise as a clinical technique.

Reliable non-invasive information on the P-R segment of the cardiac cycle would be of great clinical value for the analysis of conduction disturbances in the heart. Considerable recent effort has centered on obtaining such information from the ECG. However, practical difficulties are encountered which, so far at least, have prevented a widespread application of the method. The magnetic field of the heart offers an alternative probe which appears in principle to be better suited for recording conduction processes. Accordingly, we have obtained preliminary magnetic records using SQUID instrumentation. A very simple averaging scheme was used, employing an on-line 8 bit microprocessor. Our preliminary results confirm that the magnetic record contains clear structure, qualitatively and quantitatively consistent with conduction activity. A biphasic deflection has been observed whose onset is sharply localized in time about 40 msec prior to the onset of ventricular activity and additional activity of interest has also been noted just prior to ventricular activation. The electrical and magnetic approaches to the recording of conduction activity are assessed and it is concluded that the magnetic technique has significant advantages for this purpose.