Research Terms
Keywords
Concept Maps Decision Support Systems Health Informatics Health Information Technology Software Engineering
Industries
Researchers at the University of Central Florida, in conjunction with a physician, have created a new tool called the Dementia and Delirium Analysis Research Tool, which can conveniently and efficiently collect and analyze patient data with minimal health care provider time and involvement. This assessment improves the accuracy of delirium and dementia diagnoses as well as patient outcomes, reduces the need for expensive, repetitive testing, and can be repeated serially to document disease progression.
Technical DetailsThrough a series of questions, the Dementia and Delirium Analysis Research Tool was designed to assess a patient's mental state using the corresponding responses as a means of identification. The questionnaire consists of two main sections. The first section has a group of questions regarding the demographic information of the caretaker or caregiver and of the patient. The second section assesses the physical and mental health of the patient (e.g., questions about mood, lethargy, and sleepiness). This second section also screens for symptoms of mild dementia by asking the patient to solve math problems and to answer common knowledge questions such as "Who is the current President of the United States?"
University of Central Florida researchers have invented a decision support tool for enhancing patient care and reducing hospital readmissions related to heart failure (HF). The innovative support tool identifies and analyzes patient-centric human factors that affect hospital readmissions of patients diagnosed with heart failure. Results generated by the tool enable medical practitioners and patients to develop and use interventions that mitigate the risks of readmission. The tool analyzes items such as choice, rest, environment, nutrition, habits, activity and other human factors that affect a patient's health outcomes.
The invention comprises a novel algorithm that identifies the significance of human factors related to hospital readmissions of heart failure patients. The algorithm incorporates structural equation modeling and meta-analysis to obtain the associated probability value for individual human factors that help to reduce readmissions of heart failure patients. In one example application, the invention uses data from a systematic review and meta-analysis of clinical trial studies on heart failure hospitalization and care management strategies. The system extracts and generates relevant data, rating the significance of human factors that influence heart failure patients' knowledge, motivation, attitude, preventive practices, and health outcomes. Care managers use the results to determine the clinical interventions and practices needed to help a particular heart failure patient avoid rehospitalization.
Researchers at the University of Central Florida have developed technologies for quantitatively evaluating, tracking and managing patient medical records. With unique methodologies and statistical analyses, the technologies provide tools for assessing the strength, completeness, consistency, and accuracy of patient electronic medical records (EMRs) in one or more databases. One of the inventions also allows organizations to track chronic condition diagnoses and determine plans for staging and managing the conditions. By helping organizations identify the strengths and shortcomings of their record-keeping procedures, the UCF technologies provide the healthcare industry with a clearer path to ensuring the highest standard of care.
Technical Details
Patent ID 33510 and 34394, Method and System for Managing Healthcare Patient Record Data: These inventions provide methods and systems for managing patient EMRs in terms of completeness, consistency, and accuracy relative to established guidelines. One tool that the systems use to assess patient EMRs is the Data Completeness Analysis Package (DCAP). DCAP analyzes a patient’s medical records holistically to identify the lack of pertinent and necessary patient data. Once implemented, DCAP uses scoring algorithms and robust statistical analysis techniques to determine the completeness of individual patient records as well as aggregate patient records across health care centers and subpopulations. The system can also verify individual fields for completeness across an entire database.
Additionally, technology 34394 enables organizations to assess and compare two or more EMR databases as well as comparing the strength of a selected data field across the databases. The invention’s graphical user interface can provide remote access to a database over a network.
In one example implementation, DCAP generates visual representations (concept maps) of data in conjunction with statistical analysis. The concept mapping method works as a schema to represent stored data that users can uniformly examine regardless of the platform that originally held the data or other health care protocols that may make cross-examinations of data sets more difficult. Once developed through DCAP, the system converts the concept maps to standard data format CSV (comma-separated values) files. The CSV files are analyzed through a parser that allows the user to determine the strength of individual patient records and record-keeping throughout a particular database.
The result is a Record Score Strength (RSS) that is based upon the care provider’s input of Importance Weights (IW), along with a Patient Database Score (PDS) that defines the overall strength of the records. Using database segmentation techniques, DCAP also generates a Patient Subgroup Score (PSS) to compare subpopulations of patient records. It attains the PSS by averaging the RSS scores of the patients of interest by age, race, gender, and insurance status.
Patent ID 34565, Method and System for Managing Chronic Illness Health Care Records: The invention is a system and methods for identifying inconsistencies in the EMRs of patients with chronic illnesses. The system includes a scoring algorithm called Chronic Condition Mapping Package (CCMP) which can track chronic condition diagnoses in terms of completeness, consistency, and accuracy relative to established guidelines.
In one example use, the system first converts native EMR data into a standardized comma-separated values (CSV) data format using concept mapping to identify how data is stored. It then extracts data relevant to chronic condition diagnoses, including encounter level data, subjective narrative data, and objective data. The encounter level data can identify the stage of a chronic illness at each encounter to determine if and how the chronic illness has changed. Afterward, it analyzes the data relative to data translators, such as an ICD-10 chronic condition identifier. Once it processes the data, the system uses a complex scoring algorithm to identify and compare scores that incorporate the algorithm's aims. The scoring algorithm enables organizations to compare patient records, subpopulations, and databases, both at a chronic condition level and across chronic conditions. Scores form the basis for improvement in chronic condition data capturing and improving healthcare delivery for patients. The system also generates alerts to inform users when a pre-existing chronic illness is not subsequently identified or if a subsequent diagnosis indicates that the condition has changed by some predetermined amount.
Partnering Opportunity
The research team is looking for partners to develop the technology further for commercialization.
Stage of Development
Working prototypes available via test environment.
The University of Central Florida invention is an agent-enabled deep-learning algorithm to forecast COVID-19 positive cases with a greater degree of accuracy. The invention provides an agent learner architecture and methods to boost the capabilities of deep learning algorithms for accuracy in forecasting challenging pandemic scenarios like COVID-19. The technology also has potential applications with data-heavy sets for analysis (for example, financial markets).
Partnering Opportunity
The research team is seeking partners for licensing and/or research collaboration.
Stage of Development
Prototype available.
The University of Central Florida invention is a method and system for predicting the completeness of data in telehealth systems and capturing the true sentiments of patients. It increases the possibility of improved healthcare outcomes and enables healthcare providers to offer a greater quality of care overall.
Telemedicine, or telehealth, is one of the newest innovations in medical technology, enabling practitioners to communicate with their patients over the phone, video conferencing, or chat. However, clinical data and sentiments are often not reflected in a practitioner’s analysis and diagnosis of patients. The UCF prediction system resolves the data incompleteness in telehealth. By using telehealth natural language processing (NLP) and AI-enabled systems, the system processes the sentiments of patients and the data incompleteness found in conversations.
Partnering Opportunity
The research team is seeking partners for licensing, research collaboration, or both.
Stage of Development
Prototype available.
