Home / Technology / AI & IT / Departments of Defense developed Cloud-based biosurveillance ecosystem to warn of coming pandemics

Departments of Defense developed Cloud-based biosurveillance ecosystem to warn of coming pandemics

The threats of chemical, biological, radiological, nuclear and explosive (CBRNE) hazards continue to advance. CBRN weapons are some of the most indiscriminate and deadly weapons in existence today, with capability to affect large population in wide geographical area and in short time. The release of Chemical, Biological, Radiological and Nuclear (CBRN) materials, whether deliberate or accidental, may have the potential to cause serious harm and severe disruption to the delivery of vital public services over a wide geographical area.


In a world of rapid transit, global trade, and instability, the United States faces increased threats of intentional acts of bioterrorism, naturally occurring outbreaks of infectious disease, and accidental exposure to biological hazards. As evidenced by outbreaks of Ebola, Middle East Respiratory Syndrome, Zika, avian influenza, and Rift Valley Fever, infectious diseases emerge and spread quickly. A catastrophic biological incident, such as a terrorist attack with a weapon of mass destruction (WMD) or a naturally occurring pandemic, could cause thousands of casualties, weaken the economy, damage public morale and confidence, and threaten national security, says DHS.


While there is a growing torrent of data that disease surveillance could leverage, few effective tools exist to help public health professionals make sense of this data or that provide secure work-sharing and communication. Meanwhile, our ever more-connected world provides an increasingly receptive environment for diseases to emerge and spread rapidly making early warning and collaborative decision-making essential to saving lives and reducing the impact of outbreaks.


The Departments of Defense and Homeland Security are developing a system which lets epidemiologists scan the planet for anomalies in human and animal disease prevalence, warn of coming pandemics, and protect soldiers and others worldwide. Digital Infuzion’s previous work on the Defense Threat Reduction Agency (DTRA)’s Biosurveillance Ecosystem (BSVE) built a cloud-based platform to ingest big data with analytics to provide users a robust surveillance environment.


After the 2009 H1N1 pandemic, the Assistant Secretary of Defense for Nuclear, Chemical and Biological Defense indicated “biodefense” would include emerging infectious disease. In response, DTRA launched an initiative for an innovative, rapidly emerging capability to enable real-time biosurveillance for early warning and course of action analysis.

The US Department of Defence’s (DoD) Defense Threat Reduction Agency (DTRA) is developing a cloud-based biosurveillance ecosystem (BSVE) to help identify incongruity in human and animal diseases. Collaborating on the programme are government agencies, including the Joint Science and Technology Office for Chemical and Biological Defense (JSTO-CBD), the Joint Program Executive Office for Chemical and Biological Defense (JPEO-CBD) and Homeland Security’s National Biosurveillance Integration Center (NBIC).


DTRA scientist Dr Christopher M. Kiley said: “The BSVE is a virtual, customisable, collaborative system that uses commercial and government technologies to aggregate and analyse data streams. “The BSVE ingests and uses large data streams such as open-source social media feeds, RSS feeds from news organisations and blogs, disease ontologies, de-identified diagnostic results, historic outbreak data, zoonotic data and non-health data.” The ecosystem incorporates analytic applications with a user-friendly interface to provide near-real-time modelling, analyses and visual results.


Analytic applications and user-designed apps in the BSVE use the aggregated data streams to provide near-real-time modeling, analyses and visualized results, Kiley said. The BSVE provides automated, intelligently suggested data, tools and analyses, and a user-friendly interface with modern collaboration and reporting features.


It also supplies automated data and tools to identify aberrations in disease signals through machine learning and natural-language processing algorithms. The BVSE will be supported by a number of sources, including the World Health Organisation (WHO), the UN Food and Agriculture Organisation (FAO), the World Organisation for Animal Health (OAI), and the programme for monitoring emerging diseases (ProMED). The programme follows the US DoD’s directive for chem-bio defence missions to research emerging infectious diseases.


Data Science, Analytics and Collaboration for a Biosurveillance Ecosystem

Through competitive prototyping, DTRA selected Digital Infuzion to develop the platform and next generation analytics. This work was extended to enhance collaboration capabilities and to harness data science and advanced analytics for multi-disciplinary surveillance including climate, crop, and animal as well as human data.


We integrated over 170 global One Health data sources using cloud-based automated data ingestion workflows that provide unified access with data provenance. We used modular automated workflows to implement data science including Natural Language Processing (NLP), machine learning, anomaly detection, and expert systems for extraction of concepts from unstructured text. A first of its kind ontology for biosurveillance permits linking of data across sources. This ontology allows users to rapidly find all relevant data by looking at semantic relationships within and across data sets having varying quality, types, and usages to understand the best, most complete indicators of impending threats.


We applied the following principles to the development of data science tools: 1) mathematics should be fully automated and operate ‘under the hood’ without need for user intervention; 2) ‘At-a-Glance’ visualizations should summarize Information, draw attention to key aspects and permit drill down into underlying data; 3) data science analytics and tools need to be validated with real-world data and by disease surveillance experts and 4) secure collaboration capabilities are essential to biosurveillance activities.


The platform now provisions integrated One Health information. Data sources were harmonized and expanded, along with historical information, to better predict and understand biothreats. These include global social media, human, plant, animal, and weather data. An Analyst Workbench delivers logical, intuitive and interactive visualizations enabling disease surveillance professionals to identify critical, predictive information without extensive manual research.


To speed disease surveillance workflows, the workbench generates suggestions to the user on their current work. Anomaly detection to alert to potential developing disease events employs fully automated analytics to conduct over 43 million calculations daily for more than 500 diseases in over 170 data sources, distilling this into a table that ranks the most significant anomalous increases that may indicate an outbreak and warrant investigation.


A predictive disease modeling tool based on current and historical data uses fuzzy logic to identify the likeliest outcome, even early in an outbreak when there is much uncertainty about the disease and its characteristics. A complex automated workflow identifies health-related topics that are trending in Twitter and evaluates their severity using novel lexicons and new reactive sentiment analysis. Searches use the ontology to gather all relevant information and are supported by the most advanced NLP with custom surveillance rules to provide succinctly extracted information. This alleviates the need for extensive reading by identifying exactly which data is needed and extracting key concepts from it. Intuitive methods of visual representation, interactive displays, and drill-down capabilities were leveraged in all analytics for rapid understanding of results.


Finally, we added a software development kit to enable third party developers to continuously enhance the platform capabilities by adding new data sources and new analytic apps. This allows the platform to be adapted for specific needs and to keep pace with new scientific and technical discoveries and has resulted in over 50 analytic apps.


New analysis tools ensure the BSVE supports a One Health paradigm to best inform public health action. Digital Infuzion and DTRA first introduced the BSVE to the ISDS community at the 2013 annual conference SWAP Meet.


CWMD Integrated Biosurveillance effort

CWMD recognizes that men and women serving on the front lines of homeland security need to know as early as possible if they are at risk of exposure to a biological agent of concern. They also need to understand quickly the significance of the risk so that they can take action to protect themselves and others. For example, emerging disease threats in wildlife or the environment may exhibit features that indicate a risk for evolution to a significant outbreak of disease that ultimately could spread rapidly from person to person, resulting in a global pandemic.


Additionally, Department of Homeland Security (DHS) Countering Weapons of Mass Destruction Office
(CWMD)  will incorporate a broader scope of targets into Biofeeds. Biofeeds is a DHS-developed, open-source biosurveillance information technology (IT) tool that uses machine learning, natural language processing, and other advanced analytics components. CWMD will add more biological as well as chemical and radiological/nuclear threat domains to Biofeeds, expanding the system’s early detection/early warning capabilities to meet the priorities of CWMD better.


The sheer size and complexity of worldwide WMD data require that analysts employ advanced tools to support their efforts. A variety of tools already exist or are in development, such as Biofeeds, DTRA’s BSVE, the Defense Advanced Research Project Agency’s Sigma, Analyst Notebook, and others, to support the analysts.


The CWMD Integrated Biosurveillance effort will advance development and implementation of technologies, policies, and procedures that allow for a broad exchange of data and information. Enabling shared access to data will enhance the biosurveillance capabilities of CWMD and its partners, while also maintaining the security and privacy of the exchanged data. The program understands the expected value of expanding data access to the NTC, as well as to DOD and other agencies, to improve biological threat detection and analysis.


CWMD efforts will contribute to a layered CWMD and pandemic information architecture. The architecture allows capabilities and limits of each individual system to complement each other. It also will generate a comprehensive threat picture that includes food, agriculture, and veterinary threats as part of a One Health approach.


To identify these potential threats, the integrated biosurveillance effort is part of a CWMD structure that maintains a staff of SMEs in the areas of bioterrorism; biological, chemical, and radiological threats; intelligence; emergency response; and other areas. This staff informs biosurveillance analyses daily and serves as reachback to DHS operators in the field.




References and resources also include:



About Rajesh Uppal

Check Also

Quantum-Powered AI: Unlocking the Potential of Quantum Computing in Artificial Intelligence

In the ever-evolving landscape of technology, certain combinations have the power to redefine industries. Quantum …

error: Content is protected !!