Research

Parisa Rashidi is the director of the “Intelligent Health Lab” (i-Heal). Her area of research includes developing machine learning and artificial intelligence methods for solving healthcare problems. More specifically, she is working on: (1) transforming patient care in the Intensive Care Unit by developing autonomous monitoring tools using advanced machine learning techniques, (2) Developing intelligent tools for monitoring cognitive and mental conditions of community-dwelling patients. She has forged a highly collaborative research program across campus to address these problems, collaborating with several departments at College of Medicine including nephrology, anesthesiology, clinical neuropsychology, and aging. Her research is supported by National Science Foundation (NSF), National Institute of Health (NIH), state grants, and internal grants.

Google Scholar page

Interests

  • Machine Learning
  • Natural Language Processing
  • Intelligent Health Systems
  • Biomedical Data Science

Research Projects

  • Intelligent ICU

    Pervasive sensing and Artificial Intelligence for Critical Care

    Our long-term goal is to sense, quantify, and communicate patient condition in an autonomous, precise, and personalized manner. We plan to build the foundation of an intelligent ICU by developing and validating pervasive context sensing, precise context inference, and situation-aware context communication using novel sensing and artificial intelligence technologies. This project is carried out at Intensive Care Units (ICU) at UF Health hospitals (UFH). Several sub-projects include:

    1. Patient Activity Recognition
    2. Facial Expression Analysis, including pain and anxiety
    3. Environment Characterization (noise, light, visitors)
    4. Delirium Characterization
    5. Using Gaming, virtual reality, and mixed reality in the ICU

  • Trajectory Prediction in Hospitalized Patients

    Predicting the status of patient throughout their hospital stay

    We expect to facilitate early interventions by precisely predicting a patient’s clinical trajectory using high-resolution data and using accurate deep learning models.

  • Mental Health & AI

    Using Artificial Intelligence Techniques for Mental Health Therapy

    The goal of this project is to develop  natural language processing and machine learning techniques to provide personalized treatments to customize and individualize online mental health treatment, the Intelligent Counseling System (ICS). More specifically, we focus on several problems:

    1. Therapy recommendation
    2. Language polarity detection
    3. Cognitive distortion detection

  • Characterizing Postoperative Pain Signatures

    Finding Good Temporal Postoperative Pain Signatures

    This project is a result of collaboration with Dr. Ptarick Tighe. It examines how postoperative pain scores change with respect to time using machine learning and advanced data science techniques such as shapelets and deep learning techniques.

  • In-Community Health Monitoring

    Health monitoring using mHealth technology

    Our long term goal is to develop an ecologically momentary assessment tool for capturing patient status in real time in the context of momentary physiologic and psychometric data. We have developed an ecologically momentary assessment framework using Samsung Gear S smartwatch to assess patient reported outcomes, as well as to automatically measure physiological and psychological parameters, especially in older adults. This project is carried out in collaboration with UF Institute on Aging.

  • Postoperative Risk Assessment

    Integrating data, algorithms and clinical reasoning for surgical risk assessment

    This project is in collaboration with Dr. Azra Bihorac. It examines how surgical risk can be assessed using machine learning and advanced data analysis techniques.

  • Postoperative Cognitive Evaluation

    Presurgical Cognitive Evaluation via Digital clockfacE drawing

    This project is in collaboration with Dr. Patrick Tighe and Dr. catherine Price. It examines how deep learning and digital technology can be used to assess cognitive function in hospitalized patients.