- Platt, L.S. First Inventor: Patent: Resilient Inference System for Performance Safety (RISPS)
Utility Patent No PCT/US2021/17507, USPTO, August 2022
Predicts Infection Risks and Tests Infection Moderation Strategies in Inpatients Healthcare Environments
This Human-in-the-Loop (HITL) approach to applied Artificial Intelligence predicts infection-causing risks in inpatient healthcare environments and assesses the effectiveness of infection moderation strategies. One in 20 hospital patients contracts a healthcare-associated infection; 1.6-3.8 million healthcare-associated infections occur annually in long-term care facilities. These infections increase the need for secondary treatments and account for $28-33 billion in excess health care costs annually. Healthcare-associated infections are often preventable when appropriate intervention strategies are applied. Insights on infection-causing risks in the inpatient environment and assessments of potential moderation strategies can enable infection control teams to apply intervention procedures quickly and effectively.
Researchers at the University of Florida have developed an artificial intelligence-driven predictive process model that assesses infection-causing risks and tests infection moderation strategies in healthcare environments to support infection control team decision making.
HITL artificial intelligence that provides insights to infection control team decision-making, enabling them to implement effective strategies for moderating infection risks in inpatient settings
- Evaluates risk associated with individual components of an inpatient healthcare environment, allowing infection control teams to implement targeted risk moderation strategies
- Predictively models infection moderation strategy outcomes, identifying the most effective strategy prior to implementation
- Provides environment-specific risk assessments and predictions, enabling infection control teams to apply strategies that will be effective in a specific inpatient environment
- Uses a translatable framework, making this approach versatile and broadly applicable to assessing infection risks in other systems such as public transportation or university campuses
Healthcare-associated infections are dangerous and costly, but can be preventable when appropriate intervention strategies are applied. This trademarked artificial intelligence system, the Resilience Inference System for Performance Safety (RISPSTM), is a data-driven, decision support tool that uses risk analysis and resilience assessment to generate performance safety outcomes that provide information on the infection risk associated with individual components of the inpatient healthcare infrastructure. The predictive algorithms it is composed of help to assess the efficacy of potential infection moderation strategies within the context of that specific healthcare environment.
- Platt, L.S. First Inventor: Provisional Patent: Prevention Through Design Environmental
Material Intelligence Center (PtD-EMIC) No. T18480US001 [222107-8785], June 2021
Provides Performance Characteristics of Interior Material, Including Resistance to Bacteria, Fungi, and MRSA
This decision-support matrix of interior material performance characteristics moderates infection spread in multiple environments, ranging from hospitality to hospital care and more. Infections spread through surface transmission are common in many environments including hospitals and long-term health care environments. These infections are often preventable with an appropriate, executed strategy, but they account for $28-33 billion in excess health care costs annually. Key insights on infection-causing risks in the built environment can enable infection control teams to quickly intervene and reduce risk, but this information is not currently centralized or quickly accessible.
Researchers at the University of Florida have developed a decision-support matrix of interior material performance characteristics related to reducing bacterial bioburden,. The platform allows users to conduct basic keyword searches or use deep learning tools to extract material performance data that informs design decisions to mitigate infection risks.
This decision-support tool informs design decisions related to material choice to mitigate infection risks in the built environment
- Platform allows users to conduct keyword searches, providing material performance data that informs design decisions
- Supports use of artificial intelligence, enabling investigators to uncover patterns in interior material performance
- Can be used to produce environment-specific risk assessments, providing informational support to infection control teams
This matrix of interior material performance characteristics informs design decisions to mitigate infection risks in the built environment. This platform supports the use of artificial intelligence technologies including machine learning and data mining to reveal common patterns in interior material performance used for designing safety-critical environments such as hospitals. The platform will also be searchable using keywords and can quickly provide material performance data. Interior material performance data are based on a hierarchy of technical performance measures, including robustness, recovery, graceful extensibility, and sustained adaptability.