9 Medical Applications of Ultrawideband Radar

X-ray, computed tomography (CT), magnetic resonance imaging (MRI), and imaging methods can provide internal body images for medical diagnoses. All these have problems in terms of hazards from exposure to ionizing radiations or high and dangerous magnetic fields. These systems have special installation requirements and high initial and operating costs. This chapter presents ways to use ultrawideband (UWB) radar for medical imaging and diagnostic purposes. The combination of fine spatial resolution and reflection from any change of dielectric properties of media can provide some degree of internal body imaging without exposure to ionizing radiations or high magnetic fields.

Some commercial development of medical radar has already occurred. The Lawrence Livermore National Laboratory (LLNL) has pioneered several UWB radar applications. The U.S. Army Combat Casualty Care Program has developed a remote vital signs monitor for soldiers to wear in combat. Sensiotec, Inc., now markets a UWB radar monitoring system that goes under the patient’s mattress to remotely measure heart and respiration rates. PneumoScan, Inc., has undertaken commercial development of a LLNL UWB radar device to detect life-threatening internal bleeding in the thorax. Other companies have started development of UWB radar imaging systems for detection of intracranial hemorrhage. Further work could produce radar systems for detecting and imaging breast tumors without exposure to ionizing radiation and physical discomfort of conventional mammography. Though radar-based systems may not have the full capabilities of X-ray, CT, MRI, etc., they can provide a diagnostic imaging system at a much lower cost. This can facilitate the introduction of valuable diagnostic imaging systems into small and remote clinics, which cannot afford conventional imaging systems. As Dr. Samuel Johnson said, “Dictionaries are like watches. The worst is better than none at all, and the best don’t run true.” The same advice applies to medical diagnostic systems.

9.1.2 Chapter Overview

Section 9.2 summarizes the electrical properties of human and animal tissues. Successful medical imaging depends on the contrast of different types of radiations into tissues. Systems such as X-ray units measure attenuation of the radiation passing through the body. Medical radar systems will depend on the contrast or ratio of dielectric properties of different types of tissues to reflect radio frequency (RF) waves.

Section 9.3 presents examples of UWB radar measurement of physical movement in heart and respiration rate. The Moscow Aviation Institute Patient Monitoring Radar illustrates the principles. The Army developed vital signs monitor, and Sensiotec Virtual Medical Assistant™ provide practical examples of developed systems.

Section 9.4 discusses diagnostic tools for detecting pneumothorax and imaging inside the skull for intracranial hemorrhage and strokes.

Section 9.5 covers developmental work in medical radar imaging and microwave tomography for detecting breast cancer and strokes.

9.2 Electromagnetic Waves and Human Tissue

9.2.1 Introduction

This section describes the living tissue media where UWB medical imaging radars will operate for physiological measurements and imaging. Animal tissues have a higher conductivity than other natural materials but still remain subject to penetration by UWB signals. Chapter 5 presented the basic theory and some data on the permittivity and permeability of building materials for comparison.

This section will summarize the electrical properties of human tissue as background for further development of medical measurements and imaging systems. Successful penetration of materials depends on the proper selection of frequencies. Good imaging depends on the sufficient differences in the electrical properties of different types of tissues to create a reflecting interface surface.

9.2.2 Air Force Research Laboratory Gabriel Tissue Characteristics Database

9.2.2.1 Broad Spectrum Relative Permittivity and Conductivity of Tissues

During the 1990s, the U.S. Air Force Research Laboratory hired Camelia and Sami Gabriel of the Kings College in London, UK to prepare a database on electrical characteristics of biological tissues. They measured the relative permittivity and conductivity of many types of tissue over a frequency range from 10 Hz to 20 GHz using the instruments in Table 9.1. Their report contained a detailed discussion of the compensation methods used to determine the measured values of permittivity and conductivity [1].

They reported how the measurement techniques and associated instrumentation gave a random reproducibility of about 1% across the frequency range. Measurements included standard samples of uniform composition. They pointed out that the biological tissues show a wide variation in their structure or composition, which affects their dielectric properties. The spread in measured values ranged from ±5% above 100 MHz to ±15 at the lower end of the frequency scale.

TABLE 9.1
Relative Permittivity and Conductivity Measurement Instrumentation

Frequency Range	Instrumentation	Interface
10 Hz-10 MHz	HP4192A impedance analyzer	Open-ended coaxial probe
300 kHz-3 GHZ	HP 8753C network analyzer	Open-ended coaxial probe
130 MHz-20 GHz	HP 8720 network analyzer	Open-ended coaxial probe

Source: Gabriel, C. and Gabriel, S., Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies, U.S. Air Force Report AFOSR-TR-96.

The Gabriels collected electrical characteristic data by measuring freshly excised animal material, from cows and freshly killed sheep, and human autopsy materials. Human skin and tongue characteristics come from in vivo measurements. They used tissues taken within 2 h of death for animals and within 24-48 h for human tissues. Impedance analyzer materials required cubic pieces of at least 5 cm on each side. Because of sample size problems, they could not measure some materials at low frequencies.

To evaluate the variation of animal and human tissues, they ran a series of comparison tests of human and ovine tongue, tendon, and small intestine tissues. They observed that the variation of tissue properties within a species may exceed the variation between species [1]. However, the measurements gave a good estimate of the expected range of permittivity and conductivity values between different types of tissues and how the values varied across a broad frequency range. Figures 9.1 through 9.3 show the results of comparative tissue measurements. Note how all the cases show same patterns, that is, relative permittivity decreases with frequency, whereas conductivity increases with frequency.

9.2.2.2 Electrical Characteristics of Tissues at 500 MHz

The Gabriel reports included a table showing the relative permittivity and conductivity of various animal tissues at 500 MHz in the region of most interest to biological UWB radar designers. Tables 9.2 and 9.3 provide an idea about the relative contrasts between the different types of media UWB medical radars would encounter during remote sensing and imaging [1]. Figure 9.4 shows how the tissue properties vary with frequency.

9.2.3 Reflectivity of Biological Tissues

The difference in dielectric properties of different tissue types means that

Each tissue will have a different propagation velocity through it due to the different impedance z₀ values.
Part of the incident RF energy will reflect back to the source from any junction of two different types of tissue with different dielectric properties.
Imaging will depend on the contrast between tissues resulting from the difference of characteristics impedances and reflection from those interfaces.

Images

FIGURE 9.1
Comparison of the relative permittivity and conductivity (S/m) of sheep and cow tendon tissue over a frequency range of 1 MHz-20 GHz. (Adapted from Gabriel, C. and Gabriel, S., Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies, Figure 3, Ref [9-B], Final Report for the Period 15 December 1994-14 December 1995, Prepared for AFOSR/NL Bolling AFB DC 20332-0001, June 1996, AL/OE-TR-1996-0037, http://niremf.ifac.cnr.it/docs/DIELECTRIC/Report.html; USAF Research Lab report.)

Images

FIGURE 9.2
Comparison of the relative permittivity and conductivity of sheep and human small intestine tissue over a frequency range of 1 MHz-20 GHz. (Adapted from Gabriel, C. and Gabriel, S., Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies, Figure 4 Ref:[9-B], Final Report for the Period 15 December 1994-14 December 1995, Prepared for AFOSR/NL Bolling AFB DC 20332-0001, June 1996, AL/OE-TR-1996-0037, http://niremf.ifac.cnr.it/docs/DIELECTRIC/Report.html; AF Research Lab report.)

Images

FIGURE 9.3
Comparison of the relative permittivity and conductivity (S/m) of sheep and human tongue over a frequency range of 1 MHZ-20 GHz. (Adapted from Gabriel, C. and Gabriel, S., Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies, Figure 4, Ref:[9-B], Final Report for the Period 15 December 1994-14 December 1995, Prepared for AFOSR/NL Bolling AFB DC 20332-0001, June 1996, AL/OE-TR-1996-0037, http://niremf.ifac.cnr.it/docs/DIELECTRIC/Report.html; AF Research Lab report.)

9.2.3.1 Reflection Coefficient

We can determine the reflection coefficient from the impedance that depends on the relative permittivity ε_r and start from basics by recognizing the impedance of free space as

z_{0} (space) = \sqrt{\frac{μ_{0}}{ε_{0}}}, (9.1)

$z_{0} (space) = \sqrt{\frac{μ_{0}}{ε_{0}}}, (9.1)$

where μ₀ indicates the permeability of a vacuum at 1.257 × 10⁻⁶ H·m⁻¹ or V·A⁻¹S·m⁻¹ and ε₀ the permittivity of a vacuum value 8.854 × 10⁻¹² F·m⁻¹ or A·V⁻¹S·m⁻¹. This gives a free space value of z₀ (space) = 377 Ω. For any medium, we can estimate the impedance based on the relative permittivity, so that

z_{m e d i u m} = \sqrt{\frac{μ_{0}}{ε_{r} ε_{0}}} = \frac{z_{0} (s p a c e)}{\sqrt{ε_{r}}} = \frac{377}{\sqrt{ε_{r}}} Ω . (9.2)

$z_{m e d i u m} = \sqrt{\frac{μ_{0}}{ε_{r} ε_{0}}} = \frac{z_{0} (s p a c e)}{\sqrt{ε_{r}}} = \frac{377}{\sqrt{ε_{r}}} Ω . (9.2)$

To estimate the reflection coefficient at the interface to two different tissues, we can apply one-dimensional (1D) transmission line theory from time domain reflectometry (TDR). In the case of a transmission line, any discontinuity or change in the medium characteristic impedance causes part of the incident wave to reflect. The impedances of the two media determine the reflection coefficient:

ρ = \frac{Z_{n} - Z_{m}}{Z_{n} + Z_{m}}, (9.3)

$ρ = \frac{Z_{n} - Z_{m}}{Z_{n} + Z_{m}}, (9.3)$

TABLE 9.2
Dielectric Properties of Body Tissues at 500.00 MHz

Images

TABLE 9.3
Dielectric Properties of Head Tissues at 500.00 MHz

Type Tissue	Relative Permittivity	Conductivity (S/m)	Density (kg/m³)	Impedance (Ω_
Average brain	48.417667	0.626460	1030.0	54.18005
Average skull	17.447731	0.177212	1850.0	90.25512
Average muscle	57.324669	0.843827	1040.0	49.79326

Source: Gabriel, C. and Gabriel, S., Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies, Final Report for the Period 15 December 1994-14 December 1995, Prepared for AFOSR/NL Bolling AFB DC 20332-0001, June 1996, AL/OE-TR-1996-0037, http://niremf.ifac.cnr.it/docs/DIELECTRIC/Report.html.

Images

FIGURE 9.4
Thoracic tissue electrical characteristics. (a) Relative permittivity of thoracic tissues versus frequency compiled by C. Gabriel. (Adapted from Gabriel, C., “Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies,” Brooks Air Force Technical Report AL/OE-TR-1996-0037, Appendix D, June 1996.) (b) The different relative permittivities of each tissue type sets up detectable reflections of EM waves for range measurements and imaging. (Adapted from Paulson, C.N. et al., “Ultra-wideband radar methods and techniques of medical sensing and imaging,” SPIE Int. Symp. Optics East, October 10, 2005, UCRL-CONF-216016; Visible Human Project, U.S. National Library of Medicine, National Institutes of Heath, September 2003, http://www.nlm.nih. gov/research/visible/. With permission from SPIE © 2005 and J.T. Chang and C.N. Paulson of LLNL.)

where the first transmission media has a characteristic impedance Z_m and the second media Z_n as shown in Figure 9.5a and b. The ratio of the characteristic impedances Z_m/Z_n determines the amount of energy reflected as shown in Figure 9.5c [6].

Rewriting the equations relating characteristic impedance and permittivity as

\begin{matrix} Z_{m} = \frac{z_{0}}{\sqrt{ε_{r m}}}, \\ Z_{n} = \frac{z_{0}}{\sqrt{ε_{rn}}}, \end{matrix} (9.4)

$\begin{matrix} Z_{m} = \frac{z_{0}}{\sqrt{ε_{r m}}}, \\ Z_{n} = \frac{z_{0}}{\sqrt{ε_{rn}}}, \end{matrix} (9.4)$

where ε_rm, ε_rn indicate the relative permittivity of the media m and n as shown in Figure 9.5a. Substituting into the reflection coefficient gives a more direct relationship in terms of the relative permittivities ε_rm and ε_rn for the respective media, so

ρ = \frac{\sqrt{ε_{r m}} - \sqrt{ε_{r n}}}{\sqrt{ε_{r m}} + \sqrt{ε_{r n}}} . (9.5)

$ρ = \frac{\sqrt{ε_{r m}} - \sqrt{ε_{r n}}}{\sqrt{ε_{r m}} + \sqrt{ε_{r n}}} . (9.5)$

Note that the reflection coefficient value may vary from -1 to +1. The plot of Figure 9.5c shows the reflection coefficient absolute value for simplicity. When the characteristic impedances of both media have the same value, then the RF wave travels through without reflection. Any variations in the media characteristic impedance result in small reflections and depend on the receiver sensitivity for detection. The reflection coefficient from any given media interface sets the contrast and discrimination limit on any microwave-based imaging system.

Images

FIGURE 9.5
Reflection of radio waves between media. (a) Transmission line model for reflection from one media to another. (b) When an electromagnetic wave passes from one medium to another part of the incident energy reflects and part passes through. (Adapted from Alexandrov, O., Transmission Coefficient Optics, November 26, 2007. Open source: Wikipedia Commons.) (c) The reflectivity coefficient ρ depends on the impedances of the media. No reflection occurs when the two media have the same characteristic impedance Z_n = Z_m.

9.2.3.2 Conclusions on Microwave Radiation Tissue Characteristics

The tables of relative permittivity and conductivity presented here make a strong case for using UWB materials-penetrating radar technology for sensing and differentiating between body organs. Later sections will present examples of medical remote sensing and imaging applications. A precise knowledge of the tissue characteristics will help to develop and improve biological radar imaging and tomography.

9.3 Heart and Respiration Rate Measurement with UWB Radar

The tissue-penetrating qualities and fine spatial resolution of UWB radar provide the capability for remotely measuring heartbeat and respiration. This section will discuss the principles and performance of a remote (2 m range) system and show a commercial version for use in a special bed pad. We will start by examining the patient monitoring radar (PMR) developed by the Moscow Aviation Institute (Russia) and the Industrial Technology Research Institute (Taiwan) to show the principles and system capabilities. The U.S. Army Vital Signs Monitor will show a device for use by combat medics, and the Sensiotec Virtual Medical Assistant™ provides a system for hospitals and home care.

9.3.1 Remote PMR

Professor Igor Immoreev of the Moscow Aviation Institute (Russia) developed a PMR for remotely measuring the respiration and the heartbeat rate in hospital settings. Dr. Teh-Ho Tao of the Industrial Technology Research Institute (Taiwan) used this radar for supervision of newborn children in a hospital in the city of Tajbej, Taiwan. This UWB radar can measure movements as small as 0.1 mm made by the thorax during respiration (external) and heartbeat (internally) in a motionless person at distances of 3-3.5 m.

9.3.1.1 Technical Description of PMR

Figure 9.6a shows the PMR circuitry and antenna removed from its case. The probing signal waveform in Figure 9.6b provides the fine spatial resolution and body-penetrating qualities needed for remote measurements in patients. The PMR has a strobe range receiver feature that enables accurate measurements in rooms with many static and moving objects within the effective range. Table 9.4 summarizes the specifications of the PMR that can measure heart rhythm (HR) from 0.5 to 5 Hz (3-300 beats/breaths/min).

Figure 9.7 shows the PMR subsystems including the antenna, pulse generator, analog-to-digital converter (ADC), receiver, and power supply. It can connect to a personal computer through a standard USB connection for operational flexibility.

The PMR sits about 2 m over the patient. In each pulse repetition period, the pulse generator sends two short UWB pulses similar to the ones in Figure 9.6b, which follow each other with an interval determined by the range to the body. The first short pulse enters Key 1, the RF power amplifier, and the receiver amplifier. The RF power amplifier sends the pulse to the antenna; after the pulse radiates, both keys change state. The second short pulse from the pulse generator goes to the phase detector of the receiver. The interval between the first and second short pulse determines the reflected signal range by letting the receiver accept only signals from prescribed range from the radar to the patient. The antenna directivity creates a working measurement area about 0.30 m deep and 1 m diameter on the patient’s bed. This eliminates reflections from objects outside the bed.

Images

FIGURE 9.6
The PMR developed by the Moscow Aviation Institute and the Industrial Technology Research Institute of Taiwan. (a) The basic PMR unit removed from its case. (b) The radar signal UWB waveform. (Immoreev, I. and Tao, T-I., “UWB radar for patient monitoring,” Figures 1 & 2, IEEE A&E Syst. Mag., November 2008, 11-18, © 2008 IEEE.)

TABLE 9.4
UWB Patient Monitoring Radar Technical Specifications

Images

The pulse reflected by the preset range object returns through the antenna and low-noise amplifier into the receiving channel of the phase detector, which has two separate channels (two quadratures). Receiver signals go into the phase detector in phase, but the reference channel signals have a 90° phase shift. This means that the object-reflected signal will not appear in the low phase sensitivity area of the phase detector. Signal from both output channels enter the storage unit, which both amplifies the signals and stores them. The ADC digitizes the stored quadrature signals and transfers them to the personal computer for further processing.

Images

FIGURE 9.7
PMR block diagram showing the major subunits. This system uses a range strobe receiver to limit return signals to a 1 m diameter and 0.3 m long volume on the patient’s body and bed. The strobe feature eliminates reflections from other objects and gives the UWB radar an advantage over monitoring methods such as Doppler radar. (Immoreev, I. and Tao, T-I., “UWB radar for patient monitoring,” Figure 3, IEEE A&E Syst. Mag., November 2008, 11-18, © 2008 IEEE.)

The return signal from the PMR storage unit has a modulation with heartbeat and respiration on the same waveform. After conversion to digital format, the personal computer has signal processing algorithms to separate the heartbeat and respiration components for display and further use as shown in Figure 9.8. This can provide an important monitoring capability for patients with respiratory problems.

9.3.1.2 Clinical Tests on PMR Heart Rate Measurement

9.3.1.2.1 PMR Test Objectives

The developers tested the PMR to determine its effectiveness by long-term monitoring of respiratory, cardiac, and motion activities. They evaluated the PMR performance by comparing the output with standard methods of measuring heart rate such as electrocardiograms. Postoperative centers of two Moscow hospitals provided a place to monitor patients recovering from cardiac and blood vessel operations.

9.3.1.2.2 Test Methods and Results

The evaluation procedure placed the radar 2 m above the patient’s thorax to collect data from the patient. The test plan included collecting and comparing PMR and electrocardiograph (ECG) data from a patient. They verified the radar data periodically in 3-5 min intervals by comparing the patient’s HR and variability of heart rhythm (VHR) measured by the radar with ECG readings. The personal computer could give an alarm if the heart or respiratory rhythms exceeded some limits.

The personal computer processed and analyzed the quadrature signals and separated the HR from the respiration and motion activity. Processing the PMR data produced the time history of the heart activity and a histogram showing the heart rate versus interval between beats. Comparison of the PMR and ECG data showed a close correlation with an average deviation of 2.8% and a correlation coefficient of 0.86.

Images

FIGURE 9.8
The PMR detects the heartbeat and movement of the patient’s thorax. Because the return includes both heart and respiration combined, the personal computer separates the two components of heart rate in the upper display and breathing in the lower display. (Immoreev, I. and Tao, T-I., “UWB radar for patient monitoring,” Figures 7 and 9, IEEE A&E Syst. Mag., November 2008, 11-18, © 2008 IEEE.)

Some minor errors appeared between the PMR and ECG measurements because they indicated different things. The PMR measured the physical movement of the heart muscle, whereas the ECG measured the electrical stimulus to the muscle. Compensating for the time difference and averaging to two heartbeats improved the performance of the PMR with respect to the ECG standard. Data averaging over several heartbeats reduced the artifact influence without distorting the picture of the functions regularity. Averaging reduced the average deviation from 2.8% to 2.5% and increased the correlation coefficient from 0.86 to 0.9.

The clinical evaluation process did long-term monitoring of 10 patients in two hospitals and collected several hundred comparative radar and ECG measurements of heart rate and variation of heart rate. They found an average deviation between PMR and ECG data of 1.52% and an average correlation coefficient of 0.915.

The Moscow clinical test results recommend UWB radar for long-term remote and contactless monitoring of patients in hospitals, homes, and other settings.

9.3.1.3 PMR Respiration Rate Clinical Tests

9.3.1.3.1 Background and Test Objectives

Premature infants can experience a syndrome called apnea of prematurity (AOP) in which the newborn stops breathing. This can cause brain damage from lack of oxygen and results in permanent injury. Premature patients require AOP monitoring to assure their health and safety for a long period of time, whether they remain in the hospital or at home.

As recounted in Section 9.3.1.2.2, PMR evaluations in Moscow hospitals demonstrated the effectiveness for measuring heart rates. Because the PMR can detect both heart and breathing rates, the developers conducted a second set of respiration tests at the Chang Gun Children’s Hospital in Taiwan by measuring the respiration rates of AOP patients.

9.3.1.3.2 Respiration Measurement Test Methodology

The developers demonstrated the PMR capability to remotely measure respiration rates by comparing the radar measurements with standard impedance pneumography (IMP) measurements. Note that all current pneumographic devices require contact and electrical measurements or mechanical sensing, for example, electrodes, and flexible-inflated tube. Testing followed all the normal consent and safety protocols. Table 9.5 describes the patients and results of the tests. During the test periods, the personal computer received digitized UWB and IMP breathing signals and stored them in two files. The editing process removed segments containing noises caused by excessive body movements such as crying or caused during feeding. The analysis calculated the subject’s respiration rates that came from averaging every 10 s of data to determine the breaths/minute.

9.3.1.3.3 Respiration Measurement Test Results

The raw breathing movement data show a one-to-one correspondence between the measurements taken by the two methods. To compare the PMR and IMP performance, the researchers calculated the kernel density distributions and determined the level of agreement between the two measurement techniques that lies within the 95% confidence level according to the Bland-Altman statistical method. The two measurement methods appear statistically equivalent in these tests. Table 9.5 shows the level of agreement of the two measurement methods.

TABLE 9.5
Statistical Agreements of PMR and IMP Breathing Rate Measurements

Images

9.3.1.4 Conclusions on Evaluation of UWB Patient-Monitoring Radar

The hospital demonstrations show that the UWB PMR can detect heart and thorax motion to accurately measure heart and respiration rates. The return signal contains both measurements superimposed on one another. Data processing can separate the two for use in physiological monitoring in hospital and other situations. The PMR can accurately measure respiration and heart rates at a distance up to 3.5 m. Heart rate measurements had a better than 90% correlation with electrocardiogram measurements. Respiration rate measurements had a confidence level of 95% with respect to IMP respiration.

9.3.2 Commercialization of UWB Radar Vital Signs Monitors

The U.S. Army Medical Research and Material Command developed an LLNL UWB radar concept into a vital signs monitor. Sensiotec, Inc., has developed and now sells a UWB radar-based patient-monitoring systems. This section will briefly describe these applications and suggest some additional ideas.

9.3.2.1 U.S. Army Vital Signs Monitor

Remote measurement of heart and respiration rates generally requires attaching a set of electrodes to the body and carrying a remote wireless link. The LLNL developed the micropower impulse radar (MIR) and subsequent patents for medical applications including a body monitoring and imaging apparatus and method. Note that the MIR and the PMR described in Section 9.3.1 use essentially the same principles. Chapter 6 of Ultrawideband Radar Technology and McEwan’s MIR-related patents contain comprehensive descriptions of the MIR [8–10].

The U.S. Army’s Combat Casualty Care Research Program worked with LLNL to develop the MIR into a vital signs monitor for monitoring of front-line combat soldier and casualty assessment. The monitor shown in Figure 9.9 fits under soldiers’ clothing and body armor to measure their heart and respiration rates. The device was developed from a need to make measurement of vital signs in chemical warfare environments with the patient in a protective suit [11].

The remote vital signs monitor has many potential applications for use by soldiers in combat, operators of large and fast or dangerous machines (heavy trucks, trains, aircraft, spacecraft, etc.), and patients in ambulances, medevac flights, and field hospitals. A variation of this idea could go into automobiles to watch the driver’s state of health.

9.3.2.2 Sensiotec, Inc., Virtual Medical Assistant™ Patient Monitoring System

The Sensiotec, Inc., Virtual Medical Assistant™ system acts like a full-time private nurse and can continuously watch individual beds or an entire ward from a single station located 50 feet down the hall or miles away. The UWB radar in the bed sensor panel measures and records a patient’s heart rate and respiration. Load cells can monitor the patient’s movement in real time. The software looks for patterns and changes and can send information to pagers with room number, name, and alert type. Data from each bed goes into patients medical records and can have wide availability to personnel. The estimated price of $7000 (February, 2012) makes this a cost-effective solution to improve the overall hospital care and efficiency [12,13].

Images

FIGURE 9.9
The U.S. Army Combat Casualty Care Research Program (CCCRP) has developed a vital signs monitor based on the LLNL MIR body monitoring UWB radar. It can continuously monitor chest and heart movements through soldier’s clothing and body armor. (Adapted from Convertino, V.A. et al., Advanced Capabilities for Combat Medics, U.S. Army Institute of Surgical Research, Fort Sam, Houston, TX, September 1, 2004.)

The Virtual Medical Assistant™ system shown in Figure 9.10 includes the bed sensor panel (BSP), repeater base station (RBS), central base station (CBS), and ward monitoring station (WMS). These pieces connect together through standard wireless information exchange systems using a proprietary 900 MHz communication between BSP and RBS and Zigbee between RBS and CBS.

The BSP monitors the patient’s vital signs and movements. This 22 in. (56 cm) and 1 in. (2.5 cm) thick panel goes under the patient’s mattress below their thorax. It contains the UWB motion sensor with four transmitting and receiving antenna arrays. A communications module connects the panel to the nearest RBS and the total patient monitoring system. Table 9.6 summarizes the Sensiotec Virtual Medical Assistant™ technical characteristics. The signal has sufficient strength to penetrate the mattress material, sheets, and any clothing on the patient, which implies an effective range of about 3-5 ft. (90-160 cm) [13].

9.4 UWB Radar for Thoracic and Intracranial Trauma Diagnosis

9.4.1 Background

Lieutenant General (Dr.) George Peach Taylor Jr., the Air Force Surgeon General, said that “ . . . three top combat killers are (1) extremity hemorrhage, (2) tension pneumothorax, and (3) airway problems.” Emergency physicians and medical technicians need a quick way to determine the extent of life-threatening injuries. The time from injury to diagnosis and treatment can mean the difference between life and death.

Images

FIGURE 9.10
The Sensiotec Virtual Medical Assistant™ patient monitoring system measures heart rate, breathing rate, and patient movement. The complete set includes: (a) The 22 in. (56 cm) square by 1 in. (2.5 cm) thick BSP with a UWB radar and mechanical motion sensor, four transmit and receive antenna arrays, and communications module. (b) The RBS receives BSP signals for relay to the (c) CBS for connecting to the (d) WMS. (Courtesy of Sensiotec, Inc., Preventa vital, © 2011, http://www.sensiotec.com/products/vital.php.)

TABLE 9.6
Bed Sensor Panel Characteristics of the Sensiotec, Inc., Virtual Medical Assistant™

Images

This section describes the LLNL developed UWB radar applications for detecting pneumothorax and intracranial hemorrhages. All these system evolved from the original MIR developed by McEwan in 1996. The tissue-penetrating and fine range resolution capabilities of the MIR can quickly warn of life-threatening conditions where blood accumulates inside the thorax and between the skull and brain.

9.4.2 UWB Radar for Pneumothorax Detection

9.4.2.1 Pneumothorax Background

Life-threatening tension pneumothorax results when a lung collapses from a tumor, infection, inhaled foreign object, medical treatments, lung disease, severe trauma, or a break in a blister on the lung’s surface. This condition means that air has accumulated between the lungs and chest wall (pleural space), as shown in Figure 9.11a and b. Air pressure in the pleural space can collapse the lung and cause severe chest pain, shortness of breath, and death by pressing on the heart. Accurate diagnosis of pneumothorax requires either X-ray, CT, or MRI, which require some kind of hospital or clinical setting. Emergency medics need a way to quickly diagnose pneumothorax because the time between injury and treatment determines the outcome. LLNL developed the prototype pneumothorax detectors shown in Figure 9.11c. and licensed the technology to PneumoSonics, Inc., for commercial development.

Images

FIGURE 9.11
Pneumothorax means that a lung has been collapsed inside the rib cage. (a) Reading a chest X-ray to detect pneumothorax takes a trained physician. (Adapted from X-ray by James Heilman, Wikipedia: Pneumothorax.) (b) A CT scan showing a cross-section through a collapsed lung. (Adapted from CT Scan: Clinical Cases, Wikipedia: Pneumothorax. <en: User:Clinical Cases> http://clinicalcases.blogspot.com/2004/02/tension-pneumothorax.html, August 5, 2006 (Mar 21, 2011, en.wikipedia.org/wiki/en:Creative_Commons, Creative Commons,//creativecommons.org/licenses/by-sa/2.5/deed.en, Attribution-Share Alike 2.5 Generic license.) (c) The LLNL developed handheld pneumothorax detector can provide a quick diagnosis by taking UWB radar measurements through points on the thorax. (Adapted from LLNL, “Fast detection of a punctured lung,” LLNL Science and Technology Review, October 2007, https://www.llnl.gov/str/Oct07/pdfs/10_07.1.pdf. With permission from C.N. Paulson and J.T. Chang, © 2005.)

The LLNL MIR radar provided the sensor for measuring the depth of internal organs with respect to the chest wall. LLNL built a prototype pneumothorax detector and licensed the technology to PneumoSonics, Inc., for development and production as the PneumoScan® Pneumothorax detector. This section presents a technical overview of the PneumoScan® based on company-provided information and the company’s patent application [14,17].

9.4.2.2 PneumoScan® Pneumothorax Detector

The PneumoScan® uses the UWB pulses to noninvasively detect pneumothorax. The handheld device shown in Figure 9.12a has two components.

A probe unit connected to the main control contains an antenna for sending and receiving the reflected radar pulse.

FIGURE 9.12
The PneumoScan® pneumothorax detector. (a) The PneumoScan® pneumothorax detection system can measure the position of the chest wall and lung tissue using UWB radar to determine the extent of injuries. (Used with permission from PneumoSonics, Inc., © 2010.) (b) Medical personnel apply the PneumoScan® to points in the front and side of the chest as shown. Signal processing algorithms can determine the presence of pneumothorax conditions without the use of X-rays or CT scans. (Adapted from LLNL, “Fast detection of a punctured lung,” LLNL Science and Technology Review, October 2007, https://www.llnl.gov/str/Oct07/pdfs/10_0Z1.pdf; Wilder, S. et al., Non-Invasive Pneumothorax Detection and Apparatus, U.S. Patent Application Publication (10) Pub. No.: US 2010/0222663 Al, September 2, 2010; http://www.medical-diftributors.com/files/uploads/Product-Information-Sheet-PneumoSonics.pdf.)
A control unit contains the power source for the circuitry and sensor along with a processing system to analyze incoming data needed to detect the absence or presence of pneumothorax.

A high-speed acquisition and processing module in the control acquires data in real time and analyzes the reflected pulses. The unit can provide a yes/no indicator or other graphical data based on the measurements taken at the eight points shown in Figure 9.12b.

9.4.2.3 PneumoScan® Operating Theory

The PneumoScan® pneumothorax detector works by transmitting UWB radar pulses in to the body points shown in Figure 9.12b and receiving the returns. Each change of tissue will produce a radar reflection. Figure 9.13 shows the block and functional schematic diagrams of the electronics.

The PneumoScan® operator takes a series of readings at each of the eight points on the thorax shown, which in turn produces the radar returns shown in Figure 9.14a. The data processor collects and averages a series of returns, collects average time slot data for each increment indicating the delay (range) for each of the collection points as shown in Figure 9.14b. Pneumothorax detection results when the collected data exceed the expected values as shown in Figure 9.14c. Figure 9.15 shows the data collection and evaluation of algorithm. Decision points S10 and S12 determine pneumothorax for the right and left sides. A built in evaluation system warns the operator in the event failure or questionable readings [17].

Images

FIGURE 9.13
The PneumoScan® pneumothorax detector makes impulse radar range measurements to determine the space between the chest wall and lungs. (a) A block diagram of the PneumoScan® device showing the major subsystems. (b) A simplified schematic showing the major components. (Adapted from Wilder, S. et al., Non-Invasive Pneumothorax Detection and Apparatus, U.S. Patent Application Publication (10) Pub. No.: US 2010/0222663 Al, September 2, 2010.)

9.4.2.4 Demonstration of PneumoScan® Effectiveness

PneumoSonics, Inc., tested the PneumoScan® in hospital emergency departments on patients with possible chest trauma before an intervention by the clinical staff. The clinical staff then evaluated each patient according to the hospital protocol using either chest X-ray or CT scan. Table 9.7 summarizes the hospital test results.

Images

FIGURE 9.14
The PneumoScan® pneumothorax detector process. (a) The device collects radar returns from eight body points. (b) The data processor averages the multiple returns for each sample interval. (c) The evaluation process compares the samples for each location with expected standard deviations from each location. Measurements exceeding a threshold for each location indicate the presence of pneumothorax. (Adapted from Wilder, S. et al., Non-Invasive Pneumothorax Detection and Apparatus, U.S. Patent Application Publication (10) Pub. No.: US 2010/0222663 Al, September 2, 2010.)

Images

FIGURE 9.15
The PneumoScan® uses the data processing algorithm to measure, evaluate, and provide an indication of pneumothorax. The decision blocks S10 and S12 compare averaged measurement from each location with the expected distance from the chest wall to the lungs to determine the presence of pneumothorax and drive an indicator display. (Adapted from Wilder, S. et al., Non-Invasive Pneumothorax Detection and Apparatus, U.S. Patent Application Publication (10) Pub. No.: US 2010/0222663 Al, September 2, 2010.)

TABLE 9.7
Summary of PneumoScan® Clinical Test Results

Total patients	53
Sides examined	106
Overall pneumothorax detection accuracy	91%
Overall pneumothorax location accuracy	85%
False positives	4

Source: PneumoSonics, Inc. Welcome to PneumoSonics, http://www.pneumosonics.com/.

9.4.3 Radar Detection of Intracranial Hemorrhages

9.4.3.1 Intracranial Hematoma Physiology and Detection Methods

9.4.3.1.1 Physiology of Intracranial Hematoma

A strong blow to the head can cause the brain to impact the skull and produce a bruise or hematoma. This causes blood to collect between the skull and brain as shown in Figure 9.16. The resulting pressure on the brain can cause an increasing headache, vomiting, drowsiness and progressive loss of consciousness, dizziness, confusion, pupils of unequal size, weakness in limbs on one side of the body, and increased blood pressure. Left untreated, the intracranial hematoma may cause symptoms such as lethargy, seizures, and unconsciousness. This life-threatening condition requires immediate recognition and medical treatment [18].

9.4.3.1.2 Intracranial Hematoma Detection Methods

Successful intracranial hematoma treatment requires discovery within the “golden hour” after the occurrence. Medical responders and facilities need a way to quickly evaluate patients for intracranial hematoma without X-ray, CT, or MRI scanning equipment. A search of the alternative solutions indicates two approaches: near infrared light absorption and UWB radar impulse measurements.

The InfraScan Company introduced the Infrascanner™, a handheld scanner for detecting intracranial hematomas in 2008. The Infrascanner™ uses the differential near infrared energy absorption by normal brain tissue and blood to detect hematomas. Using the scanner on both sides of the skull can detect differences in absorption indicating a hematoma. The U.S. Navy and Romanian hospitals have purchased the InfraScan™ [19–21].

LLNL researchers developed a UWB radar-based intracranial hematoma detector and licensed it for commercial development. The following section will describe the LLNL prototype device.

Images

FIGURE 9.16
Subdural or intracranial hematoma means a collection blood inside the skull that exerts pressure on the brain. Detection methods include differential near infrared absorption and UWB radar measurements of determine displacement of the brain. (Used with permission from Mayo Foundation for Education and Research, Mayo Clinic, Intracranial hematoma, http://www.mayoclinic.com/health/intracranial-hematoma/DS00330.)

9.4.3.2 LLNL Intracranial Hematoma Detector

9.4.3.2.1 Technical Description of the LLNL Intracranial Hematoma Detector

LLNL developed the intracranial hematoma detector (IHD) using the MIR measurements to detect hematomas. The IHD uses a swept range radar to scan a space by varying the time delay from signal launch and a movable range gate. The device makes a series of range measurements of the brain surface on both sides of the skull and looks for a bilateral symmetry to identify abnormalities. In healthy patients, the returns from the right and left sides of the brain will look very similar. Differing signals indicate the presence of a hematoma. The device has the general specifications of Table 9.8.

Figure 9.17 shows the IHD concept demonstration using a RF probe on a cable. This system works by transmitting UWB radar pulses into the skull and recording the return signal.

9.4.3.2.2 Clinical Evaluation Results of the LLNL Intracranial Hemorrhage Detector

The LLNL researchers ran a series of experiments to determine if a correlation existed between the size of a simulated hematoma and the MIR sensor echo signals. Studies included a laboratory simulation and MIR scans on volunteers.

The laboratory simulation used four pig brains placed inside a human skull as shown in Figure 9.18a. Researchers added known volumes of blood and air. The UWB echo signals showed characteristic features corresponding to reflections from tissue interfaces. Signal analysis showed signal attenuation and delays on introduction of blood between the skull and brain tissue. The increased spatial offset between the brain tissues correlated well with the signal variations as shown in Figure 9.18b. As the blood volume increased, they observed a greater delay introduced by the slower propagation through blood with a higher index of refraction. The hematoma also decreased the reflected signal magnitude.

Human experiments collected MIR scans on six healthy volunteers and six patients with intracranial injuries including chronic subdural hematoma, acute subdural hematoma, acute epidural hematoma, and intracranial air (after hematoma drainage). Healthy patients showed a bilateral symmetry in returns from each side of the skull. Patients with hematoma showed similar delays and attenuations to the laboratory experiment results. A blind observer correctly classified MIR scans as healthy or containing a hematoma and correctly predicted the hematoma locations. Figure 9.19 shows the MIR, CT, and MRI results for a patient with a chronic subdural hematoma [4].

TABLE 9.8
LLNL Handheld IHD Specifications

Images

FIGURE 9.17
Clinical test version of the LLNL IHD. (a) The unit uses a RF probe applied to the skull to collect radar returns. (Used with Permission from SPIE and C.N. Paulson and J.T. Chang of the LLNL, © 2005 SPIE.) (b) Examples of radar returns in normal and abnormal conditions. (c) Blood has a higher index of refraction than other brain tissue. The accumulated blood in a hematoma will delay the response relative to normal tissue. (d) Head tissue absorption loss versus frequency. (Adapted from Haddad, W.S. and Trebes, J.E., Microwave Hemorrhagic Stroke Detector, U.S. Patent US 7,226,415,B2, June 5, 2007; Haddad, W.S. et al., Microwave Hematoma Detector. U.S. Patent US 6,233,479 Bl, May 15, 2001.)

9.4.4 Detection of Hemorrhagic Stroke

9.4.4.1 Microwave Hemorrhagic Stroke Detector (MHSD)

LLNL licensed the UWB radar hematoma detector concept for commercial development as a device for clinical settings such as emergency rooms and hospital intensive care units. This system implements the UWB radar as a tomographic system to detect bleeding inside the skull as might occur in a stroke.

9.4.4.2 Design Objectives and Benefits of the Microwave Hemorrhagic Stroke Detector

Hemorrhagic stroke (hypertensive intracerebral hemorrhage, brain bleeding) results when a blood vessel bursts inside the brain. Bleeding inside the skull will increase the pressure on the brain and can damage the brain tissues. Emergency medical technicians can use a unit like the portable LLNL IHD with the RF probe to make an initial diagnosis of intracranial hematoma. The capability of UWB radar signals to find concentrations of blood within the body opens the possibility of diagnosing other conditions inside the skull and other body regions [24]. Although all these brain injuries can have physical or mental symptoms, exact diagnosis and determination of the location requires expensive CT or MRI scanning equipment. A cheap easy alternative would give this capability to small clinics and emergency medical technicians.

Images

FIGURE 9.18
LLNL researchers demonstrated the IHD operating principle by creating a phantom hematoma and observing the UWB radar return signals. (a) The phantom hematoma made with a human skull, four pig brains, and additions of blood. (b) UWB radar signal returns for different volumes of blood added to the skull. (Adapted from Paulson, C.N. et al., “Ultra-wideband radar methods and techniques of medical sensing and imaging,” Figures 6 and 7, SPIE International Symposium on Optics East, October 10, 2005, UCRL-CONF-216016. With permission from SPIE and C.N Paulson and J.T. Chang of the LLNL, © 2005 SPIE.)

LLNL’s commercial partner intends to develop a microwave hemorrhagic stroke detector (MHSD) using UWB radar tomography system for clinics and hospitals. This section discusses the MSHD described in patents by Waleed S. Haddad and James E. Trebes [22,23].

Although intended for detecting hemorrhagic strokes, the tomographic principle could work for other parts of the body such as the abdomen and thorax. It could provide an economical way for small clinics to diagnose hemorrhagic strokes and other internal bleeding from injuries.

The MHSD can help to differentiate between ischemic and hemorrhagic strokes. Ischemic stroke results when blood flow to an area stops. A common treatment for ischemic strokes involves administering anticoagulants to promote blood flow. Hemorrhagic strokes involve bleeding and blood accumulation within the brain tissues. These require quick surgical intervention. Administration of anticoagulants in a hemorrhagic stroke can cause further damage or death. The MHSD could help save lives by quickly differentiating between hemorrhagic and ischemic conditions.

Images

FIGURE 9.19
Comparison of UWB radar scan imagery with CT axial and MRI scans of a patient with chronic subdural hematoma. (a) UWB radar scans showing difference between returns from the right and left parietal regions. (b) CT axial scan showing the hematoma (arrow upper right). (c) MRI coronal scan showing the hematoma (arrow lower right). (From Paulson, C.N. et al., “Ultra-wideband radar methods and techniques of medical sensing and imaging,” Figure 8, SPIE International Symposium on Optics East, October 10, 2005, UCRL-CONF-216016. With permission from SPIE and C.N Paulson and J.T. Chang of the LLNL, © 2005 SPIE.)

9.4.4.3 Operating Principles of MHSD

The MHSD has a UWB radar tomography system built into the special helmet shown in Figure 9.20a and b. This helmet contains a transmitter and array of receiver antennas for taking a set of measurements through the brain. The MHSD measures the UWB pulse time of flight through the skull and brain to the receiving antennas in the helmet as shown in Figure 9.20c. Any stroke or blood concentration will increase the pulse flight time and distort the waveform. Tomographic techniques and data processing provide a map of the patient’s skull and brain as shown in Figure 9.20d [22,23].

9.4.5 Summary and Conclusions

UWB radar pulses provide a way to measure the characteristics of biological tissues by measuring the RF pulse time of flight. The high index of refraction of blood helps to make hemorrhages easily distinguishable. The radar time delay and tomographic measurement techniques described here could extend to other portions of the human body. UWB radar could provide a safe and economical solution to providing diagnostic systems for emergency medical technicians in ambulances and emergency room physicians in small hospitals and clinics.

9.5 UWB Radar for Tumor Detection

The ability of the UWB radar to detect and image abnormal conditions in the skull points to other potential medical uses such as tumor detection and imaging. This section will discuss the possibilities of microwave techniques for breast cancer detection. Breast cancer presents one of the major medical problems today because of the high incidence and dangers of metastasizing to other parts of the body. Currently, breast cancer detection requires an uncomfortable X-ray procedure, special equipment and trained specialists to read the resulting mammogram. Mammography capabilities may not exist in remote locations and some patients cannot afford the procedure. A cheap, easy to use, and reliable radar-based system could save millions of lives by early detection and treatment. This section will examine microwave-based approaches to breast cancer detection, which could provide that capability.

Images

FIGURE 9.20
The MHSD applies impulse radar to tomography. (a) Helmet to hold the antenna array. (b) Transmitter and receiver antenna array for measuring radar pulse time of flight through the brain. (c) Time of flight path geometry showing a hematoma (406). (d) Reconstructed map of the patients head indicating blood concentrations. (Adapted from Haddad, W.S. and Trebes, J.E., Microwave Hemorrhagic Stroke Detector, U.S. Patent US006454711B1, September 24, 2002; Haddad, W.S. et al., Microwave Hematoma Detector, U.S. Patent US 6,233,479 B1, May 15, 2001.)

9.5.1 Tissue Contrast and UWB Radar Tumor Detection

Dr. Susan C. Hagness of the University of Wisconsin provides the rationale for applying UWB radar to breast cancer detection based on two properties of breast tissue:

Property 1: The water content of biological tissues affects the microwave interaction differently than for X-rays. Malignant tumors and normal breast tissues have almost an order of magnitude contrast for microwaves compared to X-rays or ultrasound. The large dielectric contrast gives malignant tumors a greater microwave scattering cross-section than normal tissue of similar size and shape. Figure 9.21 shows the relative permittivity and conductivity of high (muscle) and low (fat) water content tissues.

Property 2: Normal breast tissue has microwave attenuation of less than 4 dB/cm up to 10 GHz. This may permit microwave equipment with standard sensitivity and range to detect tumor as deep as 5 cm beneath the skin. Normal breasts tissue has attenuation and phase characteristics, which provide the constructive addition of wide-bandwidth back-scattered returns using broad aperture confocal-imaging techniques [25].

Images

FIGURE 9.21
The electrical properties of biological tissues depend on water content. Low water content tissue such as fat has (a) a high permittivity contrast and (b) conductivity contrast with high water content tissue such as muscle. This helps in radar detection of tumors in fatty tissue such as the breast. Plots compiled by Hagness et al. based on Gabriel et al. data. (Hagness, S.C. et al., “Two-dimensional FDTD analysis of a pulsed microwave confocal system for breast cancer detection: Fixed-focus and antenna-array sensors,” IEEE Trans. Biomed. Eng., December 1998, 45, 1470-1479, © 1998 IEEE.)

9.5.2 Radar Tumor Detection Methods

Microwave radiation has potential medical applications in detecting abnormal tissue masses inside the body, as shown in the previous sections about intracranial hemorrhage and stroke detection. Approaches to medical microwave tumor detection include radar imaging and tomography.

Radar imaging detects the radiation reflected by tumors, hematomas, etc. and measures the reflected energy and time of flight.
Tomography transmits microwave energy through the body over several paths, measures the energy transmitted in each path, and then constructs a map of the electrical properties from the received signals.

The following sections summarize radar imaging and microwave tomography methods for breast cancer detection without the use of ionizing radiation and the physical discomforts of conventional X-ray mammography. These methods can also apply to other body parts. Applying UWB signals to mammography could develop easier methods of breast cancer detection at less cost than current conventional methods and bring internal breast imaging to medical facilities in underdeveloped areas of the world [26].

9.5.3 Microwave Imaging for Medical Diagnosis

9.5.3.1 Overview

A survey of breast cancer detection patents reveals a common element of conformal microwave imaging (CMI) to construct a picture of the electrical properties of the breast tissue. If you understand the basic idea of CMI approach, then you will realize that various patents implement different methods of the same approach using either antenna arrays or moving antennas to take a series of radar measurements at different angles through the organ and then mapping the dielectric properties of the tissues.

9.5.3.2 CMI Fundamentals

CMI means using microwave reflections to construct a three-dimensional (3D) picture of the breast interior. Figure 9.22 shows the general scheme for imaging the interior of the breast or any other body part. The image formation procedure follows these steps:

Step 1: Illuminate the breast with microwaves from many locations and record the reflected signals at each location. An antenna array or a single moving antenna can do this.

Step 2: Identify the first and second skin reflection separated by a period of time. Time gate the signal by setting all data before the first skin reflection and after the second skin reflection to zero.

Step 3: Compute a first estimate of skin reflections and subtract these from each signal.

Step 4: Compute a second estimate of skin reflections from a single location from signals received in the last two locations and subtract the second reflections from the signal.

Step 5: Construct a 3D image of the interior volume from the signals showing the absence or presence of reflecting tissues.

The particular approach used in a patent depends on the antenna arrangement, for example, single moving antenna and multiple fixed antennas. Figure 9.23 shows three typical approaches to data collection using moving antennas, fixed antenna arrays, and a movable antenna array cluster. All these use the same general approach to image processing described above with variations for the radar data collection method [26–32].

9.5.3.3 Conclusions on Medical Microwave Imaging

As of the time of this book, several teams of researchers have worked on the problem of breast cancer detection by microwave imaging. Demonstrations with simulations have indicated the feasibility of the idea. Table 9.9 lists some of the patents for breast cancer detection as of April 2011. Commercial development of microwave breast cancer detection systems will mean refinements to these ideas for expense and diagnostic reliability. Medical acceptance will require clinical trials to compare the results of microwave radar detection with conventional X-ray mammography, CT, and MRI techniques. Success will depend on the demonstrated resolution and tumor detection capabilities compared with other techniques.

Images

FIGURE 9.22
CMI. (a) Imaging processes can use an array of antennas or one antenna moved to known locations. (b) General steps in the radar return data processing to build the diagnostic image. (Adapted from Winters, D.W. et al., Time Domain Inverse Scattering Techniques for Use in Microwave Imaging, U.S. Patent 7,809,427 B2, October 5, 2010.)

9.5.4 Microwave Tomography for Medical Applications

Microwave-based tomography represents another approach to medical internal imaging. The microwave hematoma detection system described earlier exemplifies the basics. Perhaps a combination of radar imaging and tomography could provide a fine resolution capability for tumor imaging and medical diagnosis.

Images

FIGURE 9.23
Some CMI antenna configurations. (a) Cavity and step moving antenna. (Adapted from Li, J. and Wang, G. Multi-Frequency Microwave Inducted Thermoacoustic Imaging of Biological Tissue, U.S. Patent 7,266,407 B2, September 4, 2007.) (b) Pillow antenna array for imaging from a multiple antenna array. (Adapted from Veen, V. et al., Space-Time Microwave Imaging for Cancer Detection, U.S. Patent 7,570,063 B2, August 4, 2009.) (c) Cluster antenna probe for hand movement over the area. (Adapted from Weide, V.D. and Warren, D., Apparatus and Method for Near-Field Imaging of Tissue, U.S. Patent 7,725,151 B2, May 25, 2010.) (d) Stationary antenna array. (Adapted from Winters, D.W. et al., Time Domain Inverse Scattering Techniques for Use in Microwave Imaging, U.S. Patent 7,809,427 B2. October 5, 2010.)

9.5.4.1 Tomography Fundamentals

Tomography means imaging the inside of an object by taking sections or by sectioning with any kind of penetrating wave.

A tomograph collects sectional data by taking a series of penetrating wave measurements through the object as shown in Figure 9.24a. The collected sectional data from each scan goes through a tomographic reconstruction process to produce a tomogram. The tomograph can collect sectional data by the three methods shown in Figure 9.24b through d, which include the following:

Rotating the object in a penetrating beam between a transmitter and receiver.
Rotating the transmitter and receiver about the object.
Using an array of transmitting and receiving elements around the object.

TABLE 9.9
Radar Breast Cancer Detection Patents

Images

FIGURE 9.24
Basic principles of tomography. (a) Tomographs pass a beam through an object and measure the received signal strength (attenuation) over a complete circle. (b) Object rotation method. (Adapted from Wikipedia, Industrial CT scanning, http://en.wikipedia.org/wiki/Industrial_CT_scanning) (c) Single beam rotation as used in X-ray CT scanners. (Adapted from Wikipedia, X-ray computed tomography [Herman, G.T., Fundamentals of Computerized Tomography: Image Reconstruction from Projection, 2nd edition, Springer, 2009]. Online: http://en.wikipedia.org/wiki/X-ray_computed_tomography#Diagnostic_use. File:Ct-internals.jpg 4/16/2011 Permission of the Free Documentation License (April 30, 2011).) (d) Multiple transmitter and receiver array scanning. (Adapted from Santhoff, J.H. Ultra-Wideband Imaging System. U.S. Pat 6,619,838 B2. July 19, 2005.)

9.5.4.2 Conventional Medical Tomography Systems

As of 2011, the principal medical tomography methods include CT, MRI, positron emission tomography (PET), ultrasonic tomography, and optical coherence tomography (OCT). All these represent established and tested technologies, which give the physician powerful diagnostic tools. Table 9.10 summarizes the current tomographic medical imaging technologies in terms of radiation form and resolution.

9.5.4.3 Microwave Tomography

Applying microwave tomography to medical applications remains a relatively new idea. A review of the patents shown in Table 9.11 indicates the use of microwaves in combination with other techniques such as ultrasound or magnetic resonance. Direct use of through tissue measurements with UWB pulses may occur in the future as shown in the proposed microwave hematoma detection system. The spatial resolution and ability to provide reliable medical information at a low cost will determine the success of any microwave tomograph for clinical use.

9.5.5 Future Directions for UWB Radar Imaging and Tomography

Perhaps a successful approach to microwave tomography would combine radar and tomographic concepts. One solution for high-resolution UWB tomography might use the direct transmitted signal combined with the forward- and side-scattered radiation from each reflector. Figure 9.25 shows the concept for a UWB side scatter imaging system.

We know that the frequency-dependent properties of the transmission media and target distort UWB waveforms as covered by Dr. Immoreev in Chapter 3. Van Blairicum described the use of the singularity expansion method for target identification by analysis of the signal returns in “Introduction to Ultra-wideband Radar Systems.” In the same book, Sheby and Marmarelis presented an approach based on higher order signal processing that demonstrated the capability of building a signature bank for identifying different types of materials by the analysis of the radar return spectrum [44].

TABLE 9.10
Some Medical Tomography Methods

Method	Radiation/Sensor	Resolution	Comments
Computer Tomography (CT)	X-rays	1 mm on a 12.5 mm slice through the body	Ionizing radiation exposure Cost $0.3 million for portable head scanner 5-10 million depending on features
Magnetic resonance imaging (MRI)	Magnets and radiation manipulate resonance of molecules in body	<12 mm thick slice Better resolution than CT	Expensive installation $1-5 million High magnetic field hazard Cannot be used if metal is present in body
Positron emission tomography (PET)	Senses radiation from injected positron emitting isotope	Neuroimaging technique. Reconstructs 3D image of distribution	Study of normal and abnormal brain function, assessing tumors, strokes, and cortical lesions Mapping of the visual cortex

Source: Tomography, The Free Dictionary, http://medical-dictionary.thefreedictionary.com/tomography.

TABLE 9.11
Summary of Microwave Tomography Patents

Images

FIGURE 9.25
The multiscatter UWB Tomograph would use the direct path, back and side scatter signals from objects as the array rotates about the object or patient. Reconstruction of the target from the direct-, forward-, and side-scattered signals could provide another approach to imaging inside objects and patients without ionizing radiation. Analysis of the received signal spectrum could provide additional data for medical diagnoses. This concept could work as a rotating or stationary array of transmit/receive antennas.

UWB signals attracted earlier researchers because of their high information content. Combining radar return spectrum analysis with imaging and tomographic techniques could provide the additional concept needed for successful UWB radar and tomographic medical applications.

9.6 Conclusions on UWB Medical Radar Applications

The materials-penetrating qualities and fine spatial resolution of UWB has potential applications to medical vital signs measuring and imaging as demonstrated by the work of Immoreev and Tao for heart and respiration rate measuring. The Sensiotec Virtual Medical Assistant™ patient monitoring system has gone on the commercial market in America.

The MIR developed by McEwan at the LLNL provided the basic technology for a military vital signs remote monitor, pneumothorax detector, and intracranial hemorrhage imager and a stroke detection system. Hopefully, all these products will eventually emerge from commercial development and certification to appear in ambulances, physician’s offices, clinics, emergency rooms, hospital wards, and home care settings. Work to date points toward possible systems for breast cancer detection using both UWB radar and tomographic techniques.

Exploiting the information-rich UWB signal through spectrum analysis can greatly improve the possibilities for medical uses of UWB radar. Combining multistatic radar and tomographic techniques could produce improvements in resolution and malignancy detection.

Acknowledgments

I wish to extend special thanks and appreciation to the following:

Professor Igor Y. Immoreev for his critical review and improvements to the description of the patient monitoring radar.
Dr. Christine M. Paulson, Dr. John T. Chang, and the LLNL for generously providing illustrations and background information on their work developing UWB radar medical applications.
Mr. Bob Purcell of PneumoSonics, Inc., for reviewing the section on the PneumoScan® pneumothorax detector and for providing permission to use the illustrations.
Efrain Gavrilovich of Sensiotec Inc. for reviewing the section on the Virtual Medical Assistant™ patient monitoring system and for providing permission to use the illustration.

References

1. Gabriel, C. and Gabriel, S. Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies. Final Report for the Period 15 December 1994-14 December 1995, Prepared for AFOSR/NL Bolling AFB DC 20332-0001, June 1996, AL/OE-TR-1996-0037. Online: http://niremf.ifac.cnr.it/docs/DIELECTRIC/Report.html

2. Gabriel, C. and Gabriel, S. Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies. U.S. Air Force Report AFOSR-TR-96.

3. Gabriel, C. Compilation of the Dielectric Properties of Body Tissues at RF and Microwave Frequencies, Brooks Air Force Technical Report AL/OE-TR-1996-0037. Appendix D, June 1996.

4. Paulson, C.N., Chang, J.T., Romero, C.E., Watson, J., Pearce, F.J., and Levin, N. Ultra-wideband radar methods and techniques of medical sensing and imaging. SPIE International Symposium on Optics East. October 10, 2005, UCRL-CONF-216016.

5. Visible Human Project. U.S. National Library of Medicine. National Institutes of Heath. September 2003. Online: http://www.nlm.nih.gov/research/visible/

6. Wikipedia, Time domain reflectometer. (Smith, P., Furse, C., and Gunther, J. Analysis of spread spectrum time domain reflectometry for wire fault location. IEEE Sensors Journal, Vol. 5, pp. 1469-11478, 2005; Furse, C., Smith, P., Mehdi, S., and Chet, M. L. Feasibility of spread spectrum sensors for location of arcs on live wires. IEEE Sensors Journal, December 2005). Online: http://en.wikipedia.org/wiki/Time_domain_reflectometer

7. Immoreev, I. and Tao, T.I. UWB radar for patient monitoring. IEEE A&E Systems Magazine, pp 11-118, November 2008.

8. Taylor, J.D. and McEwan, T.E. The micropower impulse radar (Chapter 6). In J.D. Taylor (ed.), Ultra-wideband Radar Technology, CRC Press, Boca Raton, FL. 2000.

9. McEwan, T.E. Body Monitoring and Imaging Apparatus and Method. U.S. Patent No. US005573012A, November 12, 1996.

10. McEwan, T.E. Phase Coded, Micro-Power Impulse Radar Motion Sensor. U.S. Patent No. 5, 519,400, May 21, 1996.

11. Convertino, V.A., Cooke, W.H., Salinas, J., and Holcomb, J.B. Advanced Capabilities for Combat Medics. U.S. Army Institute of Surgical Research, Fort Sam, Houston, TX, September 1, 2004.

12. Farrell, M. Nine technologies to ease pain of getting old. Forbes, July 1, 2010. Online: http://www.forbes.com/2010/07/01/technology-for-the-elderly-entrepreneurs-technology-mit_2.html

13. Sensiotec, Inc. Preventa vital. 2011. Online: http://www.sensiotec.com/products/vital.php

14. PneumoSonics, Inc. Welcome to PneumoSonics. Online: http://www.pneumosonics.com/

15. U.S. Army Military Research and Material Command. USAMRMC products portfolio. June 2007.

16. LLNL. Fast detection of a punctured lung. LLNL Science and Technology Review, October 2007. Online: https://www.llnl.gov/str/Oct07/pdfs/10_07.1.pdf

17. Wilder, S., Theobold, D., Fell, R.B. and Yoder, R.L. Non-Invasive Pneumothorax Detection and Apparatus. U.S. Patent Application Publication (10) Pub. No.: US 2010/0222663 Al, September 2, 2010.

18. Mayo Clinic. Intracranial hematoma. Online: http://www.mayoclinic.com/health/ intracranial-hematoma/DS00330

19. Infrascan, Inc. The infrascanner system. Online: http://www.infrascanner.com

20. PR Web. Infrascanner handheld brain scanner experiences significant sales growth in romania. December 2010. Online: http://www.prweb.com/releases/2010/12/prweb4878524.htm

21. PR Web. Infrascanner™ handheld brain scanner gets $2M—Addresses top cause of combat death, March 23, 2011. Online: http://www.prweb.com/releases/2010/07/prweb4272094.htm

22. Haddad, W.S. and Trebes, J.E. Microwave Hemorrhagic Stroke Detector. U.S. Patent US006454711B1, September 24, 2002.

23. Haddad, W.S., James, E.T., and Matthews, D.L. Microwave Hematoma Detector. U.S. Patent US 6,233,479 Bl, May 15, 2001.

24. Med Help. Hemorrhagic stroke. Online: http://www.medhelp.org/medical-information/show/567/hemorrhagic-stroke

25. Hagness, S.C., Taflove, A., and Bridges, J.E. Two-dimensional FDTD analysis of a pulsed microwave confocal system for breast cancer detection: Fixed-focus and antenna-array sensors. IEEE Transactions on Biomedical Engineering, Vol. 45, No. 12, December 1998, pp. 1470-11479.

26. Fear, E., Okoniewski, M., and Stuchly, M. Tissue Sensing Adaptive Radar Imaging for Breast Tumor Detection. U.S. Patent 7,454,242 B2. November 18, 2008.

27. Santhoff, J.H. Ultra-Wideband Imaging System. U.S. Patent 6,919,838 B2. July 19, 2005.

28. Weide, V.D. and Warren, D. Apparatus and Method for Near-Field Imaging of Tissue. U.S. Patent 7,725,151 B2. May 25, 2010.

29. Veen, V., Barry, D., Hagness, S.C., Bond, E.J., and Li, X. Space-Time Microwave Imaging for Cancer Detection. U.S. Patent 7,570,063 B2. August 4, 2009.

30. Li, J. and Wang, G. Multi-Frequency Microwave Inducted Thermoacoustic Imaging of Biological Tissue. U.S. Patent 7,266,407 B2. September 4, 2007.

31. Bond, J.E., Hagness, S.C., and Van Veen, B.D. Surface Identification Using Microwave Signals for Microwave-Based Detection of Cancer. U.S. Patent 7,647,089 B2. January 12, 2010.

32. Winters, D.W., Van Veen, B.D., and Hagness, S.C. Time Domain Inverse Scattering Techniques for Use In Microwave Imaging. U.S. Patent 7,809,427 B2. October 5, 2010.

33. Li, J. Breast Cancer Detection System. U.S. Patent 7,340,292 B2. March 4, 2008.

34. Santhoff, J.H. Ultra-Wideband Imaging System. U.S. Pat 6,619,838 B2. July 19, 2005.

35. Wikipedia, Industrial CT scanning. (CT Services Overview. Jesse Garant & Associates. August 17, 2010. Online: http://www.jgarantmc.com/ct-services.html; Flisch, A. et al. Industrial computer tomography in reverse engineering applications. DGZfP-Proceedings BB 67-CD Paper 8, Computerized Tomography for Industrial Applications and Image Processing in Radiology, Berlin, Germany, March 15-17, 1999; Hofmann, J., Flisch, A., Obrist, A., Adaptive CT scanning-mesh based optimisation methods for industrial X-ray computer tomography applications. NDT&E International, Vol. 37, 2004, pp. 271-2278; Noel, J. Advantages of CT in 3D Scanning of Industrial Parts. August 18, 2010. Online: http://www.3dscanningtechnologies.com/pdfs/parts.pdf; Reducing preproduction inspection costs with industrial (CT) computed tomography. Commercial Micro Manufacturing Magazine for the global micro manufacturing technology industry, August 2010. Online: http://www.micromanu.com/x/guideArchiveArticle.html?id=1695; Woods, S. 3-D CT Inspection Offers a Full View of Microparts, November 1, 2010. Online: http://micromanufacturing.com/showthread.php?t=876; Zoofan, B. 3D Micro-Tomography—A Powerful Engineering Tool. 3D Scanning Technologies. July 5, 2010. Online: http://www.3dscanningtechnologies.com/ComputedTomographyPage.php3D_Micro-Tomography_-_A_Powerful_Engineering_Tool-4/). Online: http://en.wikipedia.org/wiki/Industrial_CT_scanning

36. Wikipedia, X-ray computed tomography. (Herman, G.T. Fundamentals of Computerized Tomography: Image Reconstruction from Projection, 2nd ed., Springer Verlag, London, 2009). Online: http://en.wikipedia.org/wiki/X-ray_computed_tomography#Diagnostic_use

37. Tomography, The Free Dictionary. Online: http://medical-dictionary.thefreedictionary.com/tomography

38. Ciocan, R.M. and Jiang, H. Microwave Imaging Assisted Ultrasonically. U.S. Patent 7,725,167 B2. May 25, 2010.

39. Ciocan, R.M. and Jiang, H. Microwave Imaging Assisted Ultrasonically. U.S. Patent 2007/0015993 Al. January 18, 2007.

40. Meaney, P.M. and Paulsen, K.D. Two-Dimensional Microwave Imaging Apparatus and Methods. U.S. Pat 5,841,288. November 24, 1998.

41. Gilmore, C. and LoVetri, J. System and Methods of Improved Tomography Imaging. U.S. Patent 2011/0046907 Al. February 24, 2011.

42. Meaney, P.M., Paulsen, K.D., and Weaver, J.B. Imaging by Magnetic Resonance Adsorption, Elastography and Tomography. US 2009/0036766 A l. February 5, 2009.

43. Meaney, P.M., Paulsen, K.D., and Weaver, J.B. Imaging By Magnetic Resonance Adsorption, Elastography and Tomography. US 2006/0012367. January 19, 2006.

44. Taylor, J.D. Introduction to Ultra-wideband Radar Systems. CRC Press, Boca Raton, FL 1995.

45. Winters, D.W., Van Veen, B.D., and Hagness, S.C. Time Domain Inverse Scattering Techniques For Use In Microwave Imaging. U.S. Patent Application 20060241409, October 26, 2006.

46. Haddad, W.S. and Trebes, J.E., Microwave Hemorrhagic Stroke Detector, U.S. Patent US 7,226,415,B2, June 5, 2007.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for 9 Medical Applications of Ultrawideband Radar

Create new playlist

Sign In

Sign Up

9.3.1.2.1 PMR Test Objectives

9.3.1.2.2 Test Methods and Results

9.3.1.3.1 Background and Test Objectives

9.3.1.3.2 Respiration Measurement Test Methodology

9.3.1.3.3 Respiration Measurement Test Results

9.4.3.1.1 Physiology of Intracranial Hematoma

9.4.3.1.2 Intracranial Hematoma Detection Methods

9.4.3.2.1 Technical Description of the LLNL Intracranial Hematoma Detector

9.4.3.2.2 Clinical Evaluation Results of the LLNL Intracranial Hemorrhage Detector

Table of Contents for
9 Medical Applications of Ultrawideband Radar