People live longer. Seriously ill patients have multiple chronic diseases. Multiple pathophysiological processes and their interactions must be tracked concurrently. Because of the interactions between organ dysfunctions and drug side effects, the treatment of one organ may worsen the impairment of another organ. Keeping track of the dynamics of complex diseases, e.g., sepsis, mentally is error-prone, especially during handover between medical staff. According to a 2016 study by the CDC, each year at least 1.7 million adults in America develop sepsis.
Our goal is to turn the tide through the research on computational pathophysiology models and use these models to develop medical best practice systems.
It transforms the informal description of medical knowledge to precise software that *tracks and records* complex disease dynamics. This simplifies the disease management process for clinicians and enables software verification and validation.
It guides medical staff through the best practice in real-time at the bedside. Each recommendation from our system consists of 1) what to do, 2) the list of abnormalities that trigger the recommendation, and 3) a reference to the applicable medical best practice text. The link between patient conditions and matching best practices has been the key to a high compliance rate. The accurately recorded patient condition and treatment makes *handover* much easier.
To verify the correctness of software, we partner with Prof. Grigore Rosu to develop a physician friend executable formal specification MediK, which can automatically generate: 1) Flowcharts used in medical algorithms and 2) Formal analysis tools (powered by K framework). An implementation of the guidance systems that can interact with any GUI applications through a standard interface.
To verify the correctness of software, we partner with Prof. Grigore Rosu to develop a physician friend executable formal specification MediK, which can automatically generate: 1) Flowcharts used in medical algorithms and 2) Formal analysis tools (powered by K framework). An implementation of the guidance systems that can interact with any GUI applications through a standard interface.
Different hospitals have different clinical environments. We partner with Prof. Shangping Ren to develop assumption management tools. Our assumption management technology helps to ensure the presumed facilities and supports are the same as actual facilities and supports in different ICUs.
To ensure the medical correctness of our computational pathophysiology models and the safety and effectiveness of our guidance systems, we partner with: 1) Dr. Jonathan Gehlbach and Dr. Paul M. Jeziorczak of OSF on pediatric sepsis and COVID 19 ARDS (lung failure); 2) Dr. Priti Jani of University of Chicago Medical Center on COVID 19 cardiopulmonary arrest resuscitation and 3) with Dr. Karen White of Carle Foundation Hospital on adult cardiac arrest resuscitation.
Like GPS navigation transforming the navigation history, we shall demonstrate that AI-enhanced and computational pathophysiology model-based medical best practice guidance systems will transform clinical practice.
Funded through support from NSF grants 1329886 “Integrated Emergency CPS - Human Systems,” and NSF1545002 “An Executable Distributed Medical Best Practice Guidance System for End-to-End Emergency Care from Rural to Regional Center,” we have been working with
Carle Foundation Hospital on cardiac arrest and adult sepsis, and recently with OSF Children’s Hospital of Illinois on pediatric sepsis. Since the start of COVID, we are funded by C3.ai to integrate AI-based early warning with our guidance system technology to create:
* COVID 19 ARDS (lung failure) guidance system with OSF Children’s Hospital of Illinois, Here is the demonstration prototype
* COVID 19 Cardiopulmonary arrest resuscitation system with University of Chicago Medical Center. We are in the process of creating a demonstration prototype