Category: Data and Analytics

Responding to the COVID-19 pandemic necessitates the adoption of traditional public health measures for disease control, including rapidly tracing the contacts and locations of infected individuals to prevent further spread. Leveraging technology can accelerate the scale and speed of this essential public health measure, and many organizations have announced plans to launch or support contact tracing initiatives though their underlying technologies, workflows, and privacy architectures.

In April 2020, B.Next and In-Q-Tel convened members of the public health and technology communities for a virtual roundtable to discuss the key elements of tech-enabled contact tracing programs, potential strategies to optimize their implementation and adoption, and possible frameworks for preserving the privacy and civil liberties of Americans. View the presentations on YouTube to learn more about:

Note, these presentations were recorded June 19, 2020.

Diagnostic tests are a critical tool to contain epidemics, to support medical care, and for public health measures. Understanding when they are accurate and inaccurate is necessary for understanding which individuals have the virus, need isolation, and need their contacts traced.

Many diagnostic tests are reliable, though all are imperfect. And at a large scale, tiny errors in accuracy for single tests can aggregate into large errors if deployed without care. This is especially true when the rate of true infection in the tested population is expected to be low. For example, when testing for infection in a person who doesn’t have symptoms or a history of exposure, or when testing for a history of infection when the overall prevalence of disease for a given population is low.

This paper covers:

  • The accuracy and errors in diagnostic tests
  • How low disease prevalence can cause many false positives
  • Diagnostic testing in normal times: testing for influenza
  • Balancing errors with diagnostic needs

B.Next’s experts from healthcare, government, and industry leveraged its knowledge and expansive network to create the following high-level guide that maps sensors in commercial products to key vital signs and explores ways to capitalize on the smart products that may supplement digital health efforts in response to COVID-19.

What’s Inside:

  • Connection between health sensing techniques and relevant vital signs
  • Highlight of some challenges with existing commercial solutions
  • Catalog of current product capabilities and emerging trends for future products
  • Use cases for understanding potential ways to use sensors during contingency care

This Technology Insights Guide provides an overview of the use of commercial-off-the-shelf (not medical grade) sensor technologies that might be useful in supporting the ongoing COVID-19 outbreak response. The intent of this guide is to acquaint clinicians with the fundamental concepts related to the use of wearable sensors and to provide a possible adjunct to existing healthcare related strategies for managing the potential surge of patients with COVID-19 symptoms. Emphasis on technologies that could serve as an adjunct to conventional methods of patient monitoring are provided. This effort is not intended to convey medical guidance or provide recommendations regarding the outpatient management of presumed or confirmed COVID-19 patients. It is intended as a supplement to support the ongoing efforts of the healthcare community currently managing COVID-19.

Background – This paper reports on a November 15, 2018 Roundtable that explored opportunities and obstacles associated with applications of machine learning, deep learning, neural networks and other forms of “artificial intelligence” to bioscience, biotechnology and their application to biomedicine. The Roundtable was convened by IQT Labs, the research venture of In-Q-Tel (IQT), in collaboration with Lawrence Livermore National Laboratory (LLNL). Roundtable participants included multidisciplinary experts from industry, academia, finance and several U.S. government agencies. The discussion took place over a single day, included invited presentations from three participants, and was held on a not-for-attribution basis.

An effective response to a disease outbreak requires the rapid identification of pathogen and source.

Exploiting genomic information has become an important component of effective biothreat agent identification, characterization, and attribution. To do so, the necessary bioinformatic analyses require known genomic data against which to compare the agent’s genomic data. However, more and more genomic data is becoming privately held. To truly understand where an agent came from or important features of the agent (e.g., virulence, alternative hosts, and environmental stability), the biodefense community will likely need to leverage the genomic data that resides in these private databases. This may be especially important when a truly novel agent is discovered and near-neighbors need to be identified. Security requirements necessary for biothreat agent information or active investigations limit the direct sharing of genomic information with outside parties. Private entities are often unable to share access to their database due to privacy and legal issues. Fortunately, technology options exist that enable secure computations to be executed that fulfill data privacy requirements.

We developed the Secure Interrogation of Genomic Databases (SIG-DB) algorithm to enable the interrogation of a privately held database with a sequence of interest to determine the presence of similar sequences, without compromising the query or database information. This method was confirmed to be functional and evaluated using wild-type and in silico mutated versions of Escherichia coli and Staphylococcus aureus genomic sequences obtained from the NCBI RefSeq database.

This is the poster that was presented at the 2018 annual biothreats meeting, hosted by the American Society for Microbiology (ASM).

Genomic data are becoming increasingly valuable as we develop methods to utilize the information at scale and gain a greater understanding of how genetic information relates to biological function.  Advances in synthetic biology and the low cost of sequencing are increasing the amount of privately held genomic data.  As the quantity and value of private genomic data grow, so does the incentive to acquire and protect such data, which in turn creates a need to store and process these data securely.

This project explores the limitations, opportunities, and capabilities of secure computation techniques applied to DNA sequence comparisons. Using homomorphic encryption (a software-based encryption approach) the Secure Interrogation of Genomic DataBases (SIG-DB) protocol was developed to enable searches of databases (DB) of genomic sequences with an encrypted query sequence without revealing the query sequence to the database owner or any of the database sequences to the Querier. Our results show that the SIG-DB algorithm returns an accurate assessment of the similarity of queries to databases of interest. The computational runtime and information leakage were compared between a fully homomorphic approach using the Microsoft SEAL cryptosystem and a partially homomorphic approach using the Paillier cryptosystem.  SIG-DB is the first application that we are aware of to take advantage of locality-sensitive hashing and homomorphic encryption to allow generalized sequence-to-sequence comparisons of genomic data.

We also explored an alternative approach that uses hardware-based secure computation, specifically Software Guard eXtension (SGX), by Intel®. We were unable to complete a prototype at this time due to the immaturity of the technology.  However, our research findings indicate that SGX has the potential to enable a cloud-based secure computation system with, theoretically, minimal information leakage and similarity scoring execution times near equivalent to plaintext comparisons.  Much research remains to be done to fully understand the operational and security limitations of the system.

We briefed government stakeholders on our prototype and findings at a recent B.Next event.  Attendees expressed strong support for continued work on homomorphic encryption for secure interrogation of genomic databases.  The participants provided valuable feedback on the tool and numerous use cases they encounter that could be transformed by this approach.  Although the algorithm was developed specifically for microbial genomics comparisons, SIG-DB could be useful for a number of applications, including healthcare, human genomics, organizational collaborations, and more.

Leveraging multiple data variables is critical for effective infectious disease outbreak management. Contextual actionable data provided to decision makers is often best analyzed and conveyed visually. However, limited human resources and collaborative platforms create many challenges for the effective use of data.  B.Next executed a project to explore, evaluate, and demonstrate how and to what extent information technology capabilities might empower public health analysts with limited or no coding experience to create enhanced data visualizations during an infectious disease outbreak. Dynamic visualizations with multiple variables have proven beneficial in past epidemic management operations for situational awareness. The popular tools available to produce interactive visualizations require financial resources beyond what is often available to state and local public health organizations or coding expertise to utilize open source coding libraries such as JavaScript, R, and Python. This report outlines the methods and results of a B.Next project with Plotly, a company that creates open source tools for visualization, to further develop their existing web-based interface to create interactive cross-filtering visualizations with multivariate datasets for non-coders.  Plotly enhanced their Chart Studio tool to enable cross-filtering functionalities with nine different charts in a dashboard display. An evaluation of Plotly’s enhancement was completed by both public health subject matter experts and IQT Labs personnel with expertise in infectious disease and visualization tools. The evaluation suggests that the cross-filtering functionality within Plotly provides new capabilities accessible to non-coders. However, Plotly needs additional improvements to compete with more intuitive interfaces such as Tableau.  

We live in an era of increasingly frequent and impactful infectious disease outbreaks. Naturally occurring outbreaks will have significant regional and international security implications for the foreseeable future, given the negative impacts (i.e., death, societal disruption, and economic costs) of such events. We also live in an information era. Integrating novel and available data technologies into public health practice will improve situational awareness, help shape outbreak interventions more precisely, facilitate faster and more efficient response activities, and save lives. To realize these efficiencies, federal, state and local public health agencies need a fundamentally more aggressive and systematic adoption, use and coordination of data technologies to provide essential information for tailoring interventions during an outbreak. Current and emerging data technologies can help tackle the next epidemic.

Dissecting the Anatomy of a Pandemic

Pandemics (diseases that spread globally) are rare events that are often devastating, causing substantial mortality and economic damage. Just like hurricanes or earthquakes, efforts to understand the origins of pandemics and predict their emergence would help reduce their impact and ultimately prevent them. However, unlike earthquakes or hurricanes, our efforts to understand the causes, patterns, and origins of pandemics are only just beginning. Here I highlight recent advances in disease ecology, virology, and biogeography that move us towards these goals. I also identify five critical questions that, if answered, will greatly enhance our ability to predict and prevent pandemics.

Predicting pandemics first requires analyzing trends and common themes in their emergence

Over the past few decades we have learned a great deal about the anatomy of a pandemic. Most pandemics originate as zoonoses (diseases from animals, mainly wildlife). In fact, every one of the true pandemics of the last 50 years has either originated entirely within a wild animal species (e.g., SARS originating in bats) or contains genes derived from wild animal viruses (e.g., pandemic influenza A H1N1/09 virus). Because most pandemics are caused by viruses, this article will focus on them.

Why Do We Need Vaccines?

Smallpox, polio, measles — control of these lethal diseases is possible because of vaccines. Vaccines are not only the most effective way to eradicate an infectious disease, but are also critically important for protecting first responders and noncombatant (civilian) populations from the consequences of a bioterror event. The U.S. government has expended substantial resources to protect the nation against a potential bioterror event, creating specialized planning and preparedness units within the Departments of State, Defense, Energy, Agriculture, Homeland Security, and Health and Human Services in an e ort to comply with the World Health Organization’s (WHO) International Health Regulations. These agencies work together to accelerate progress toward a world “safe and secure from infectious disease threats” within the frame of
the recent five-year Global Health Security Agenda (2014-2018). Several federally subsidized advanced development and manufacturing production facilities in different regions of the country are capable of producing millions of doses of protein-based vaccines. Unfortunately, despite these important advances in the strategic preparedness of U.S. agencies for biodefense, vaccine design remains a significant obstacle to national biodefense. Director of Biomedical Advanced Research and Development Authority (BARDA) Robin Robinson recently stated, “We can produce vaccines faster, but we also need to make vaccines more effective”. This is particularly true for the very real threat of new pathogens, for which little is known about the critical antigenic determinants and correlates of immunity, the key parameters used in conventional vaccine design.