The Next Generation of Sensor Technology for the BioWatch Program

"March 2008 Update: The Autonomous Pathogen Detection System (APDS) described in this article was deployed on a limited basis in New York City in December 2007.[#_edn1][i]"

Introduction

Mankind is capable of misapplying many biological agents, including bacteria, viruses, and biological toxins that cause fatal and/or incapacitating diseases, with the intent to harm others. The efficacy of biological weapons rests on several unique properties. Biological agents are often not detectable by human senses, and the symptoms of an exposure often do not present themselves for several days or longer, enabling a user to deploy the agent in a clandestine fashion and escape, thus avoiding attribution. The September–October 2001 dissemination by an unknown party of sporulated Bacillus anthracis, the causative agent of anthrax, in the U.S. postal system is illustrative. Certain biological agents, such as variola major, the causative viral agent of smallpox, are contagious, and airborne transmission from human to human may rapidly spread disease in a population. Because bacterial and viral agents may multiply rapidly in a host organism, a low dose of pathogen, even a few individual cells or particles, in some cases may be sufficient for infection.

Whether they are present naturally or due to human release, the detection of pathogenic species in open environments such as air and water supplies is critical in order to provide ample warning to a populace of possible contamination, contain the human-to-human transmission of a contagious agent, reduce the number of victims of an outbreak, and provide timely treatment for those who have been infected. The onset of symptoms resulting from human exposure to a pathogen often occurs after several days and usually is not easily distinguished from more common ailments. Nevertheless, the human immune response may provide the only other means of pathogen detection available. The need to rapidly and reliably detect biological agents in the environment is especially acute due to both their lethality and prior use by terrorist groups against civilians. While there exist a number of biological sensor systems developed for military use, they are of limited utility for civilian defense purposes. An effective civil sensor system must be capable of detecting low aerosol concentrations of pathogenic species in complex urban environments subject to, for example, high levels of airborne pollution, pollens, and benign, native bioorganisms. In addition, the resolution of the sensor system should be sufficiently high in order help to determine the source and geographic distribution of the agent in question.

This issue brief describes one civil biodefense system, known as BioWatch, currently deployed in the United States to detect airborne biological agents and provide early warning of a pathogen release to local, state, and federal agencies responsible for medical response, containment, and incident investigation. After addressing the capabilities and deficiencies of the current BioWatch detection and analysis system, the brief discusses the next generation of autonomous sensors currently in development for use in BioWatch. Autonomous sensors promise to perform biological analyses with greater speed, sensitivity, and selectivity for pathogens of interest, thereby enhancing the overall effectiveness of aerosol monitoring as a national defense tool. A focus of the brief is the Autonomous Pathogen Detection System (APDS), an autonomous sensor developed by Lawrence Livermore National Laboratory (LLNL) capable of both pathogen detection and identification. This examination of APDS helps illustrate the potential capabilities of and challenges facing autonomous sensors now in development as they prepare for nationwide deployment in 2009.

Overview of Current BioWatch Program

After the terrorist attacks of 2001, the U.S. federal government made civilian biosecurity a national priority. According to the Center for Arms Control and Non-proliferation, since 2001 the U.S. has invested more than $40 billion in biosecurity-related initiatives.[1] The BioWatch program is a small part of this effort. The Department of Homeland Security (DHS) deployed the BioWatch program in 2003 to serve as an early warning system for aerosolized pathogen release by monitoring high threat urban environments for the presence of airborne pathogens. Since 2003, the Science and Technology Directorate of DHS has funded and managed BioWatch, although oversight of the program was recently transferred to the DHS Office of Health Affairs.

Ultimately, BioWatch will form one component of a more comprehensive interagency surveillance system designed to identify and respond to pathogen outbreaks. This system, known as the National Bio-surveillance Integration System, or NBIS, is in its infancy and remains under development.[2],[3] NBIS seeks to combine urban aerosol monitoring data obtained from BioWatch with information from several other sources, including medical surveillance programs such as the BioSense program managed by CDC and veterinary surveillance conducted by the Department of Agriculture. The goal of this integration is to rapidly identify anomalous levels of pathogenic agents resulting from a natural event (such as the emergence of the H5N1 influenza virus) or a deliberate attack, in order to facilitate an appropriate government response.

A number of press reports and government documents have revealed the general operational framework of BioWatch, although few specifics are publicly available due to the sensitivity of the program.[4] Briefly, BioWatch is comprised of three main components: environmental aerosol sampling using stationary monitors, sample analysis, and response to a positive result. In addition to DHS, two other federal organizations, the Centers for Disease Control and Prevention (CDC) and the Environmental Protection Agency (EPA), contribute their expertise to various aspects of the program. EPA is principally responsible for the aerosol sampling component of BioWatch. Current environmental aerosol sampling equipment is reportedly based on the Biological Aerosol Sentry and Information System (BASIS) developed by LLNL and the Los Alamos National Laboratory. These aerosol samplers now are deployed in and around at least 30 major cities, and share site locations with EPA air quality monitors. The samplers collect airborne particles, including bacteria and viral particles, onto solid filters using a vacuum system. Reports published in the general media often characterize the samplers as "sensors," a misleading descriptor since it implies that the aerosol samplers are capable of independent particle characterization. Rather, the sampler filters are manually collected at 24-hour intervals and transported to a nearby laboratory member of the Laboratory Response Network (LRN) for analysis. Established in 1999 by the CDC, the LRN is a national network of approximately 150 federal, state, and local laboratories equipped to promptly identify biological agents.[5] Upon receipt of a sample filter, the lab performs a number of analyses to characterize the particles captured on the filters. In particular, the LRN relies upon the polymerase chain reaction (PCR) to amplify and identify the genomic material (e.g. DNA or RNA) of the suspected pathogen. By using PCR, the LRN is capable of identifying minute quantities (potentially down to a single particle or organism) of a biological agent trapped in a BioWatch filter.[6] However, because PCR-based analysis relies on the presence of genetic material, BioWatch may not be capable of identifying toxins, such as ricin or botulinum toxin.[7]

BioWatch is configured to identify highly pathogenic species selected from the CDC's Category A and Category B lists of select agents.[8],[9] The functional detection limits of the BioWatch system, which are dependent on aerosol sampler design and location in addition to laboratory analysis, are not publicly known, although on multiple occasions BioWatch has proven capable of detecting low levels of harmful pathogens under real world conditions. For example, in September 2005 BioWatch detected Francisella tularensis, a Category A select agent, in the Mall area of Washington, D.C. at an environmental level considered as not dangerous to human health.[10]

A number of factors inhibit the effectiveness of the BioWatch program in its current form. BioWatch is incapable of detecting a biological release as it occurs or soon thereafter. Because filter retrieval, transport, and analysis are all performed manually, pathogen identification requires up to 36 hours from the time of particle capture, assuming a 24-hour filter collection interval.[11] In order to provide health officials and law enforcement the best opportunity to respond effectively, the time between particle capture and pathogen identification should be reduced, with instantaneous identification being ideal. Furthermore, due to the desiccating action of solid air filters and potentially long delay between particle entrapment and filter collection, captured viruses and bacteria may not be well-preserved and therefore it could be difficult to assess many important pathogen characteristics (such as viability and antibiotic resistance).[12] Perhaps most importantly, the entire collection/analysis process is labor intensive and requires substantial human resources. According to DHS, labor costs dominate the operational budget of BioWatch, and impose practical and financial limits on the number of aerosol samplers that the program is able to deploy.[13] BioWatch currently operates on an estimated annual budget of $85 million, and the annual maintenance cost of monitoring a single city is reportedly $1 million.[14],[15] BioWatch may require more resources in certain cities where DHS has recently expanded the air sampler network to achieve more effective geographic coverage. Due to such cost considerations, an expansion of BioWatch beyond major cities is unfeasible at this time.

There are indications that BioWatch has not performed as intended. For example, many biodefense experts have questioned the ability of BioWatch to detect small-scale pathogen releases.[16] According to a 2005 EPA Office of Inspector General report, air samplers are often located several miles apart and are not optimally distributed for biological agent detection.[17] Further, EPA has failed to ensure proper upkeep of aerosol sampling equipment. A 2007 DHS Office of Inspector General report disclosed that BioWatch has historically suffered significant quality control and interagency communication problems, usually due to human error.[18] The most common problems encountered in the course of two evaluations of field and laboratory operations involved the improper handling and transfer of air filters, which, given the potentially low aerosol concentrations of target pathogens, could compromise the integrity of the filters through cross-contamination and therefore result in false analytical results. DHS has taken corrective actions in response to the audit to the satisfaction of the Inspector General, but the continued reliance on human handling of sensitive filter equipment is nonetheless a weakness of the current system.

Can Autonomous Sensors Enhance BioWatch?

In principle, current concerns surrounding BioWatch, including system cost, analytical speed and accuracy, geographical coverage, and error suppression may all be addressed through the adoption of state-of-the-art detection technologies. Soon after the initial deployment of BioWatch, DHS initiated an effort to foster the creation of more advanced sensor technology in order to evolve BioWatch into a truly nationwide sensor network. In September 2003, the Homeland Security Advanced Research Projects Agency (HSARPA) issued a solicitation for research proposals aimed at the development of a new environmental biological agent sensor known as the Bioagent Autonomous Networked Detector (BAND).[19] The central aim of the BAND program is to develop fully autonomous sensors capable of continuous aerosol monitoring and sample analysis without direct human intervention. Autonomous sensors enable faster, more frequent sample analyses, significantly greater cost efficiency, and broader geographical sensor coverage.[20] In eliminating the need for manual sample collection and handling, autonomous sensors reduce significantly the risk of false analysis due to human error. The target performance requirements, ownership costs, and system characteristics of BAND established by HSARPA are demanding. For example, the sensors must be capable of continuous, completely autonomous operation for 30 days or more, while performing aerosol collection, sample preparation, and the simultaneous analysis for at least 20 Category A and Category B select agents with an extremely low false positive rate (less than one false positive in 10,000,000 analyses). In order to achieve at least a four-fold reduction in operational cost per unit versus current air samplers, the sensors themselves should cost less than $25,000 apiece with an annual maintenance cost of $10,000 (roughly $0.17 per test).[21] Finally, BAND must operate in both outdoor and indoor environments while occupying a small physical volume of roughly two cubic feet.

HSARPA established a three-phase selection process beginning in 2004 to guide the development of an autonomous BioWatch sensor. After evaluating at least 14 competing proposals during the two initial phases of the BAND program, HSARPA awarded Phase 3 BAND contracts to three private companies, U.S. Genomics, IQuum, and Microfluidic Systems, Inc., within the past year.[22] Although few specifics detailing their respective BAND systems are publicly available, it appears that the underlying technologies of all three companies focus on the characterization of pathogenic genetic material as the primary means of pathogen identification.[23],[24],[25] The companies will submit prototypes of autonomous sensors to DHS for field testing by May 2008. Whichever BAND design is selected by DHS following prototype evaluation likely will form the basis of the "Generation 3" BioWatch program. In testimony before the U.S. House of Representatives in May 2006, a DHS representative stated that pilot testing of the BAND system is to commence in 2008, followed by deployment during 2009–2012.[26],[27]

Examining the Autonomous Pathogen Detection System

Are fully autonomous, broad-spectrum biological sensors within the grasp of modern technology? If so, how might they function? The U.S. military has historically lead U.S. biological agent sensor research and development efforts, and as such has developed autonomous biological agent point detection systems such as the Biological Integrated Detection System (BIDS) and the Joint Biological Point Detection System (JBPDS).[28] According to the U.S. Department of Defense (DoD), JBPDS can simultaneously identify up to ten aerosolized biological agents, in less than 20 minutes.[29] Upon detecting aerosolized particles of biological origin, JBPDS employs tests known as immunoassays to identify the target species.[30] An immunoassay is an analytical test that uses the strong and specific interactions between antigens and antibodies to identify antigens, and is frequently used for the detection and characterization of pathogens. While the military has deployed JBPDS within the U.S. on a limited basis, the system is still under development and it is unclear at this time whether JBPDS may be readily adapted for widespread civilian use. Few details describing JBPDS sensor function and sensitivity, error rate, maintenance requirements and cost are publicly available, making such an assessment challenging.

Apart from military systems, the development of autonomous biological agent sensors intended solely for civilian use has seen significant progress over the past decade. Referring to the open literature, the Autonomous Pathogen Detection System (APDS) developed at LLNL appears to be the most extensively characterized of such systems. In contrast to JBPDS, LLNL scientists designed APDS expressly for use in civilian settings, including outdoor environments, large buildings, transit centers, and special events.[31] The mission of APDS clearly parallels that of the BAND systems under development for BioWatch. An analysis of APDS therefore provides some insight into the likely capabilities and potential limitations of the BAND systems.

APDS is a completely self-contained, podium-sized instrument that is transportable between locales. The system is capable of 7 days of uninterrupted, maintenance-free aerosol monitoring with sample analysis performed once every hour.[32] APDS performs several functions automatically: continuous aerosol collection, sample preparation, two distinct types of sample analysis (immunoassay and real-time PCR), and wireless transmission of analytical results to a control center. The aerosol collector samples up to 3,300 liters of air per minute and traps aerosolized particles in water for analysis. This method of collection already represents an improvement over the dry filter technology currently in use for BioWatch as it preserves viable organisms for later laboratory study (such as growth in culture), if desired.

The two analytical technologies employed by APDS, immunoassays and PCR-based analysis of genomic material, are widely used in biological sensor systems (although the specific configurations of such tests vary widely). Several key factors distinguish APDS: the ability to use immunoassays and PCR, a mechanism enabling simultaneous analysis for multiple pathogens, and the capability to do so without a human operator. APDS analyzes aerosol samples rapidly; sample preparation, analysis and identification occur within one hour. The system accomplishes this by testing the samples for up to 100 pathogenic species concurrently (a technique known as multiplexing) using an innovative bead-based color-coding system. In short, APDS uses color-coded beads, each color being specific to a particular pathogen, to perform the immunoassays and PCR analyses. The identity of the pathogen is determined simply by examining the color of the beads that respond during an analysis. A graphical overview of the immunoassay-based analyses is illustrated below in Figure 1, and the PCR analyses are now performed using a related strategy. By utilizing the two independent test types, APDS maintains a very low false positive rate, a critical requirement for the new BioWatch sensors outlined in the BAND requirements. On the down side, the analyses may not recognize previously unknown organisms such as naturally occurring mutant viral strains or genetically engineered bacteria, resulting in false negative results. This is a common weakness facing autonomous detection systems currently under development.

The practical utility of APDS for use in the civilian sector is unclear at this time. Under controlled, indoor aerosol release conditions at U.S. Army Dugway Proving Grounds, APDS successfully detected and identified live B. anthracis and Yersinia pestis (plague), as well as denatured botulinum toxin. [33],[34] APDS correctly identified multiple pathogens during a mixed release and also distinguished B. anthracis from other sporulating bacteria, such as B. sultilis and B. globigii. In these studies, the aerosol concentration of B. globigii detected was approximately 49 particles per liter of air; however, HSARPA calls for a limit of detection (LOD) of roughly 100 organisms captured from 18,000 liters for the BAND system.19 APDS has also been tested in "real world" environments such as airports and subways, and more recently several APDS units deployed in one urban area successfully functioned as a sensor network. According to John Dzenitis, Project Leader for the APDS at LLNL, the system is capable of operating in demanding environments with good levels of sensitivity. APDS successfully performed over 30,000 sample assays without registering a false positive, but its LOD and false negative rate (i.e. the probability it will fail to detect a biological agent[35] have not been published openly. Despite impressive results, further development may be needed before APDS or a related system is ready for deployment as part of BioWatch. According to Edward Rhyne, Program Manager for the Chemical and Biological Research and Development Section at the Science and Technology Directorate at DHS, "APDS does not meet some of the key BAND requirements," although specific shortcomings were not identified.[36] Analyzing these possible shortcomings provides insight into the challenges likely to confront the privately developed BAND systems. The major issue currently facing APDS is the cost of acquisition and operation, which remains "a challenge" in the eyes of its developers.[37] As with the current BioWatch program, the major factor limiting the geographic coverage area (and hence overall efficacy) of the next generation autonomous system will be the cost of operation of a deployed network. Therefore, identifying potential solutions to the problem of sensor operational cost is paramount if BioWatch is to be successful. One of the oft-cited critiques leveled against BioWatch and aerosol monitoring generally as a civil defense strategy is a lack of cost-effectiveness in defending against a "low probability, high consequence" pathogen release. For example, in arguing against a blanket deployment of biological sensors nationwide, JASON, a scientific advisory group, estimates that such an effort would cost $10–15 billion, which would be a huge expansion of BioWatch's current $85 million annual budget.[38]

A key strategy for limiting operational cost is maximizing the time period between scheduled maintenance sessions. For example, APDS can operate continuously for one week without human intervention, yet BAND must operate for one month or more in order to minimize labor costs. Although it is unclear at this time how the BAND systems will accomplish this, they certainly will seek to minimize the consumables necessary for analysis, including the expensive antibody- and oligonucleotude-based chemical probes used in immunoassays and PCR, respectively, by miniaturizing system components and thereby enabling a greater number of analyses per maintenance cycle. This approach would also provide a sensor product that satisfies other BAND criteria not met by APDS, most notably a small instrument footprint.

Indeed, research scientists working in U.S. National Laboratories have already developed new systems that improve upon several characteristics of APDS, such as instrument size and speed of analysis. For example, LLNL has adapted APDS's color-coded bead technology to develop a more compact sensor referred to as FluIDx.[39] FluIDx is intended to serve primarily as a rapid diagnostic instrument for respiratory viruses in hospitals, clinics, and other point of care facilities, although FluIDx could in principle be adapted for use as a continuous environmental monitor for BioWatch-type efforts. Importantly, the primarily medical application of FluIDx highlights the potential extension of technologies initially developed for biodefense applications to other key areas of public health, such as timely diagnosis of infectious disease. In addition, LLNL and Sandia National Laboratories have recently collaborated to design the BioBriefcase, a small (roughly two cubic feet) self-contained system for use in pathogen identification.[40],[41] BioBriefcase employs a similar pathogen determination strategy as APDS (multiplexed immunoassays and PCR-based assays), and makes use of microfluidics to reduce consumable use and instrument size.[42] Although it is too early to perform cost analyses of these systems, their emergence makes it clear that reductions in the use of consumables and instrument size, both ultimately relevant to instrument cost, are achievable.

Finally, accurate and reliable operation in the "real-world" is the most important criterion the BAND sensors must satisfy, yet is arguably the most challenging to achieve and difficult to precisely measure. An assessment of sensor function must involve determining a LOD, a false positive rate, and a false negative rate. The sensor's false positive rate may be determined in part through real-world operation in the absence of a biological agent, as described above for APDS. However in order to determine the LOD for a given pathogen, as well as a false negative rate, realistic field testing must be performed using the pathogen itself or an adequate simulant. For obvious reasons, it is impossible to conduct live agent testing safely in an open urban environment, yet attempting to replicate such complex environmental conditions in a more controlled setting would be costly and potentially futile. Live agent testing in an aerosol chamber provides useful information about the basic function of the sensor system, but cannot possibly capture the reality of an urban aerosol mixture. The challenge therefore facing the BAND systems is to demonstrate to policymakers that they will function, with a high degree of confidence, in myriad outdoor conditions. If an autonomous sensor establishes such confidence, it will likely find a use in civil biodefense. For example, in 2005 the US Postal Service completed installation of the Biohazard Detection System (BDS), an autonomous sensor designed to detect B. anthracis, at hundreds of its branches nationwide.[43] Clearly, as an indoor system intended to detect a single pathogen type, BDS faced a lower bar for establishing satisfactory levels of confidence prior to deployment, but its adaptation nonetheless signals a desire at the federal level to apply autonomous pathogen detection strategies.

Conclusion and Outlook

Nearly five years after its deployment, the overall utility and practicality of the BioWatch program is yet to be determined in the eyes of many. Critics point to BioWatch's current limitations, such as slow rate of pathogen detection, inability to detect pathogens in multiple sample matrices, high cost of acquisition and operation, and limited deployment as evidence that environmental monitoring is an improper strategy for detection of biological agent releases. JASON, noting the highly evolved ability of the human immune system to detect foreign invaders, advocates a pathogen detection strategy centered on medical surveillance; in their words, medical surveillance "relies on the American people as a network of 288 million mobile sensors with the capacity to self-report exposures of medical consequence to a broad range of pathogens."[44] Others propose close symptomatic monitoring of sentinel populations, such as police and firefighters, to detect pathogen release. Indeed, the development of real-time medical surveillance, in the form of government programs such as CDC's BioSense and DoD's ESSENCE, is a central thrust of current biodefense efforts.[45],[46]

In the near term, domestic pathogen detection efforts will most likely rely on medical diagnosis and government monitoring of the health and agricultural sectors for aberrant disease patterns, a general strategy known as syndromic surveillance. However, as noted earlier, medical symptoms of a biological release often do not present themselves for many days, and even then may be confused with more common and innocuous ailments. Syndromic surveillance therefore is likely to detect a biological attack, but not within a desired timeframe. The continued evolution of autonomous sensor technologies will, in the near future, lead to functional sensor systems that offer viable alternatives to syndromic surveillance. This brief has highlighted a few of such systems, yet many alternate sensor technologies are under development that may ultimately supercede these in terms of cost and effectiveness.[47] A particularly exciting research avenue is the continued miniaturization of sensor systems that now require a sizable operational footprint, such as APDS. Smaller systems will require less power and fewer consumables, reducing operational costs and enabling them to be used in more diverse settings. Whether autonomous environmental aerosol monitors will become sufficiently cost-effective to warrant widespread deployment as part of BioWatch is unclear, but such systems will play at least a limited role in the national biodefense strategy. Autonomous biological sensors also will serve functions well beyond biodefense; the same underlying analytical technologies incorporated into these systems are currently used for medical diagnostics and food- and water-borne pathogen detection and identification. In this light, investments made by the U.S. government and private industries aimed at advancing sensor technology is clearly worthwhile. Ultimately, with responsible investment in the near-term, the technologies underlying biological sensors are likely to become inexpensive enough to enable sensor use in a wide variety of public settings.

Sources:

[i]Spencer S. Hsu, "New York Presses to Deploy More Bioweapons Sensors; DHS Priority Is Development of Next-Generation Devices," Washington Post, January 9, 2008, p. A3.
[1]"Federal Funding for Biological Weapons Prevention and Defense, Fiscal Years 2001 to 2008," Center for Arms Control and Nonproliferation, www.armscontrolcenter.org .
[2] Kimothy Smith, testimony before the US House of Representatives, Committee on Homeland Security, Creating a Nationwide Integrated Biosurveillance Network, 109th cong., 2nd sess., May 11, 2006.
[3] Office of Inspector General, Department of Homeland Security, Better Management Needed for the National Bio-surveillance Integration System Program, (Department of Homeland Security, Washington, D.C., July 2007), OIG-07-61.
[4] Dana Shea and Sarah Lister, Congressional Research Service, The BioWatch Program: Detection of Bioterrorism (Washington, DC: Library of Congress, November 19, 2003).
[5] "Facts About the Laboratory Response Network," Centers for Disease Control and Prevention, www.bt.cdc.gov.
[6] Dana Shea and Sarah Lister, Congressional Research Service, The BioWatch Program: Detection of Bioterrorism (Washington, DC: Library of Congress, November 19, 2003), p. 10.
[7] Charles McQueary, testimony before the US House of Representatives, Select Committee on Homeland Security, Homeland Security Science and Technology; Budget Hearing for Fiscal Year 2005, 108th Cong., 2nd sess., February 25, 2004, p. 37.
[8] "Development Continues of U.S. Bio, Chem Sensors," Global Security Newswire, January 11, 2007.
[9] "Bioterrorism Agents/Diseases," Centers for Disease Control and Prevention, www.bt.cdc.gov.
[10] Martin Weil and Susan Levine, "Biohazard Sensors Triggered; Mall Germ Levels Likely Not a Threat," Washington Post, October 1, 2005, p. B1.
[11] Office of Inspector General, Environmental Protection Agency, EPA Needs to Fulfill Its Designated Responsibilities to Ensure Effective BioWatch Program, (US Environmental Protection Agency, March 23, 2005), 2005-P-00012, p. 3.
[12] Dana Shea and Sarah Lister, Congressional Research Service, The BioWatch Program: Detection of Bioterrorism (Washington, DC: Library of Congress, November 19, 2003), p. 10.
[13] John Vitko, testimony before the US House of Representatives, Committee on Homeland Security, Creating a Nation-Wide Integrated Biosurveillance Network, 109th cong., 2nd sess., May 11, 2006.
[14] Crystal Franco and Shana Deitch, "Billions for Biodefense: Federal Agency Biodefense Funding, FY2007–FY2008," Biosecurity and Bioterrorism: Biodefense Strategy, Practice, and Science 5(2), (2007), p. 117.
[15] Office of Inspector General, EPA, 2005, p. 6.
[16] Robert Armstrong, Patricia Coomber, Stephen Prior, Ashley Dincher, "Looking for Trouble: A Policymaker's Guide to Biosensing," Center for Technology and National Security Policy, National Defense University, June 2004, p. 44.
[17] Office of Inspector General, EPA, 2005, p. 8.
[18] Office of Inspector General, Department of Homeland Security, DHS' Management of BioWatch Program, (Department of Homeland Security, Washington, D.C., January 2007), OIG-07-22.
[19] Department of Homeland Security, Homeland Security Advanced Research Projects Agency, Detection Systems for Biological and Chemical Countermeasures (DSBCC), Research Announcement 03-01 (US Department of Homeland Security, 2003), p. 7.
[20] Edward Rhyne, Program Manager, Chemical & Biological Research and Development Section, DHS Science and Technology Directorate, e-mail correspondence with author, August 27, 2007.
[21] Edward Rhyne, Program Manager, Chemical & Biological Research and Development Section, DHS Science and Technology Directorate, e-mail correspondence with author, August 27, 2007.
[22] "IQuum, MFSI, U.S. Genomics Have Phase 3 BAND Work," Terror Response Technology Report, May 30, 2007.
[23] "Biodefense," US Genomics, www.usgenomics.com.
[24] "The Products; Liat Analyzer," IQuum, www.iquum.com.
[25] "The Complete Solution – Bioagent Autonomous Networked Detector (BAND)," Microfluidic Systems, Inc., www.microfluidicsystems.com.
[26] John Vitko, testimony before the US House of Representatives, Committee on Homeland Security, Creating a Nation-Wide Integrated Biosurveillance Network, 109th cong., 2nd sess., May 11, 2006.
[27] "Lab-in-a-Box," Department of Homeland Security, www.homelandsecurity.org.
[28] US Department of Defense, Chemical and Biological Defense Program, Annual Report to Congress, (April 2007), p. B-1.
[29] US Department of Defense, Chemical and Biological Defense Program, Annual Report to Congress, (April 2007), p. B-1.
[30] "JBPDS; Joint Biological Point Detection System," General Dynamics, www.gdatp.com.
[31] Lawrence Livermore National Laboratory, 2005 Annual Report, p. 19, www.llnl.gov.
[32] Benjamin J. Hindson, et al., "Autonomous Detection of Aerosolized Biological Agents by Multiplexed Immunoassay with Polymerase Chain Reaction Confirmation," Analytical Chemistry 77 (2005), p. 284-289.
[33] Benjamin J. Hindson, et al., "Autonomous Detection of Aerosolized Biological Agents by Multiplexed Immunoassay with Polymerase Chain Reaction Confirmation," Analytical Chemistry 77 (2005), p. 284-289.
[34] Mary T. McBride, et al., "Autonomous Detection of Aerosolized Bacillus anthracis and Yersinia pestis," Analytical Chemistry 75, (2003), p. 5293-5299.
[35] Benjamin J. Hindson, Anthony J. Makarewicz, Ujwal S. Setlur, Bruce D. Henderer, Mary T. McBride, John M. Dzenitis, "APDS: The Autonomous Pathogen Detection System," Biosensors and Bioelectronics 20 (2005), p. 1925-1931.
[36] Edward Rhyne, Program Manager, Chemical & Biological Research and Development Section, DHS Science and Technology Directorate, e-mail correspondence with author, August 27, 2007.
[37] John Dzenitis, Project Leader for the APDS, e-mail correspondence with the author, August 15, 2007.
[38] Henry Abarbanel, Steven Block, Sidney Drell, Freeman Dyson, Robert Henderson, Steven Koonin, Nate Lewis, Roy Schwitters, Peter Weinberger, Ellen Williams, JASON, Biodetection Architectures, The Mitre Corporation, February 2003, p. 5.
[39] Mary T. McBride, Benjamin J. Hindson, Steve B. Brown, "Multiplexed Diagnostic Platform for Point-of-Care Pathogen Detection," US Patent Application No. 2007/0166725, Jul. 19, 2007, pp. 1-7.
[40] "Defense Against Chemical and Biological Threats-BioBriefcase, Sandia National Laboratories, www.sandia.gov.
[41] John M. Dzenitis, William J. Benett, Raymond P. Mariella, Jr., Steven R. Visuri, Kodumudi S. Venkateswaran, "Biobriefcase," US Patent Application No. 2006/0281101, Dec. 14, 2006, pp. 2-4.
[42] "LLNL, Sandia Seek Help Developing BioBriefcase," Terror Response Technology Report, September 6, 2006.
[43] "GeneXpert Overview," Cepheid Inc., www.cepheid.com.
[44] Henry Abarbanel, Steven Block, Sidney Drell, Freeman Dyson, Robert Henderson, Steven Koonin, Nate Lewis, Roy Schwitters, Peter Weinberger, Ellen Williams, JASON, Biodetection Architectures, The Mitre Corporation, February 2003, p. 1.
[45] "BioSense; Background," Centers for Disease Control and Prevention, www.cdc.gov.
[46] "ESSENCE: Electronic Surveillance System for the Early Notification of Community-based Epidemics," Department of Defense, www.geis.fhp.osd.mil.
[47] US Environmental Protection Agency, Office of Water, Office of Science and Technology, Technologies and Techniques for Early Warning Systems to Monitor and Evaluate Drinking Water Quality: A State-of-the-Art Review (Washington, DC: EPA, August 25, 2005), EPA/600/R-05/156, p. 81.